J. Meteor. Res.  2018, Vol. 32 Issue (3): 367-379   PDF    
http://dx.doi.org/10.1007/s13351-018-7101-2
The Chinese Meteorological Society
0

Article Information

LU, Chunhui, and Botao ZHOU, 2018.
Influences of the 11-yr Sunspot Cycle and Polar Vortex Oscillation on Observed Winter Temperature Variations in China. 2018.
J. Meteor. Res., 32(3): 367-379
http://dx.doi.org/10.1007/s13351-018-7101-2

Article History

Received June 28, 2017
in final form January 10, 2018
Influences of the 11-yr Sunspot Cycle and Polar Vortex Oscillation on Observed Winter Temperature Variations in China
Chunhui LU1, Botao ZHOU1,2     
1. Laboratory for Climate Studies, National Climate Center, China Meteorological Administration, Beijing 100081;
2. Collaborative Innovation Center on Forecast and Evaluation of Meteorological Disasters, Nanjing University of Information Science & Technology, Nanjing 210044
ABSTRACT: Using the NCEP-2 reanalysis data in 1979–2015, we analyze variations in the coupled stratosphere–troposphere system and attribute them to the polar vortex oscillation (PVO) and the 11-yr sunspot cycle (SC). Subsequently, influences of PVO and SC on the near-ground temperature and extreme temperatures are diagnosed based on observations at 2419 surface stations in China over the same period. Empirical Orthogonal Function (EOF) analysis of geopotential height (GH) anomalies indicates that the first and second EOF modes together can explain nearly 50% of the total variance and they have different driving sources, active periods, and regions. The first EOF mode mainly represents variation characteristics of the polar vortex, and its active periods appear in late winter. It is found that a weakened polar vortex (larger amplitude in the positive time series of the first mode) corresponds to lower daily mean, minimum, and maximum temperatures and more frequent cold nights and days. This cooling effect mainly occur in northeastern China. The second EOF mode is closely related to the SC, and its major active periods are late autumn and early winter. The results reveal that strong solar activity (larger amplitude in the positive time series of the second mode) leads to cooling effects in northern China through accelerating seasonal transformation of the stratospheric circulation and enhancing intensity of the subtropical westerly jet in the upper troposphere and lower stratosphere. The near-ground temperature is lower than usual, especially for daily mean and minimum temperatures. The number of warm nights and days is significantly reduced, and cold nights and days become more frequent. Therefore, the first and second EOF mode time series of GH anomalies can be used as indices of PVO and solar activity, respectively; and can provide indications of winter cooling processes in China.
Key words: 11-yr sunspot cycle     polar vortex oscillation     extreme temperature indices     cooling process    
1 Introduction

In recent decades, there have been remarkable changes in observed temperature and in the occurrence of extreme temperature events on both the global and regional scales (Alexander et al., 2006; Caesar et al., 2006; Brown et al., 2008; Hartmann et al., 2013). In China, several influential extreme cold events have occurred in winter, such as the strong snowstorm during winter 2007/08 (Zhou et al., 2011) and the super cold surge during winter 2015/16 (Jiang et al., 2016; Sun et al., 2018). Extreme weather in different forms, including freezing rain and heavy snowfall caused by cold surges, can severely affect regional transportation, human health, agriculture, and so on. Therefore, understanding why these variations in observed winter temperature occur and distinguishing between the contributions of various types of external forcing and internal variabilities are important to disaster prevention and mitigation and to future projection of extreme weather and climate.

Solar radiation is the main energy source for the earth and is one of the most important external forcing for the earth system. Observations have indicated a typical sunspot cycle (SC) with an 11-yr period, and during this cycle the main difference in radiative energy occurs in the ultraviolet (UV) range. The UV radiative energy increases by about 6%–8% between solar maxima and minima in a solar cycle, while across the full spectrum of solar radiation, the enhanced energy is small (0.1%) during strong solar cycle years (Chandra and McPeters, 1994). The effects of solar radiation are mainly concentrated in two areas (Andrews et al., 1987): the near-surface troposphere, which absorbs longwave radiation; and the stratospheric ozone layer, which absorbs UV radiation. The differences in UV radiation caused by the SC can affect the coupled stratosphere–troposphere system by changing the temperature distribution in the ozone layer. Labitzke (1987) first proposed the relationship between the Arctic stratospheric temperature at 50 hPa and the SC, and with advances in the reanalysis data and availability of observations at high altitudes, Labitzke et al. (2001, 2003, 2005) further indicated that in different phases of the Quasi-Biennial Oscillation (QBO) of the equatorial zonal wind, solar activity exerted different effects on atmospheric circulation through changing the atmospheric dynamical processes. Kodera (2002, 2003) and Kodera and Kuroda (2002) proposed that in boreal winter, the strength of solar activity could affect the spatial structure of the North Atlantic Oscillation (NAO), leading to apparent climate and weather anomalies in the European areas.

Recently, more and more studies started to focus on the solar cycle influences and indicated that solar activity plays an important role in global weather and climate through different ways (Gray et al., 2010; Kodera et al., 2016; Matthes et al., 2017; Xiao et al., 2017; Zhao et al., 2017). Based on dynamic diagnosis of planetary waves and simulations using coupled climate models, Ineson et al. (2011) and Lu et al. (2017) indicated that downward reflection may act as a “top-down” pathway by which the effect of solar UV radiation on the upper stratosphere can be transmitted to the troposphere and affect the European and North American regions. In East Asia, it was also proved that the 11-yr SC was able to modulate the structural changes of Arctic Oscillation (AO) as well as its influences on tropospheric circulations indirectly (Liu and Lu, 2010; Chen and Zhou, 2012). Wang et al. (2015) further indicated that the influences of solar activity on East Asian winter climate were asymmetric in different phases of the solar cycle.

The polar vortex is the most important circulation system in boreal winter, with dominant influences on the global weather and climate. Thompson and Wallace (1998) defined the first Empirical Orthogonal Function (EOF) mode of sea level pressure as the AO, after which many studies investigated the influences of the polar vortex on the troposphere by focusing on AO-related anomalies (Park et al., 2011; Chen et al., 2013; Li et al., 2014; Park and Ahn, 2016). Baldwin and Dunkerton (1999) proposed that spatial structures similar to the AO not only existed in the troposphere but also extended upward to the stratosphere, especially in polar areas, and that a strong cold vortex occupied the entire polar stratosphere–troposphere system. The concept of the Northern Hemisphere Annular Mode (NAM) was proposed, and the NAM index was employed to investigate the variations in the polar vortex and the related interactions between the stratosphere and troposphere (Baldwin et al., 2003; Cai and Ren, 2007; Lu et al., 2016).

As mentioned above, a number of studies have shown important influences of both the 11-yr SC and polar vortex oscillation (PVO) on the atmospheric circulation, while most of them conducted the analyses based on some specific indices (such as the 10.7-cm solar radiation flux and NAM index) and focused on the respective impacts of these two factors. In this study, we define two new indices—the PVO index and the SC correlation (SCC) index based on the significant variation features of the stratosphere–troposphere system, and diagnose the connections between these features and near-ground winter temperature and extreme temperature indices derived from 2419 surface meteorological observation stations in China. We then analyze the reasons behind these relationships.

In this paper, Section 2 introduces the data and methods. Section 3 examines variations of the stratosphere– troposphere system through EOF analysis of geopotential height anomalies. Section 4 reveals the correlation between the SCC and PVO indices and the observed near-surface temperature anomalies in China in winter during 1979–2015. Finally, Section 5 presents the conclusions and discussion.

2 Data and methods

To investigate the variation characteristics of the coupled stratosphere–troposphere system, the NCEP-2 reanalysis data (Kanamitsu et al., 2002) are used to conduct the EOF analysis of the geopotential height anomalies. The NCEP-2 data have 17 vertical levels covering 1000 to 10 hPa and a T159 spectral resolution (Kanamitsu et al., 2002). Here, we employ a 144 × 73 (longitude × latitude) grid of monthly mean data from January 1979 to December 2015. Differences between the original data and their respective mean climatology are defined as anomalies to analyze the annual variations and their influences on the observed changes in winter temperature in China.

To describe the SC, we select the 10.7-cm solar radiation flux as a representative index, because it has a good consistency with the evolution of SC. Furthermore, in this study, we apply the monthly mean data of this solar activity index over 1979–2015 provided by the NOAA National Centers for Environmental Information (https://www.ngdc.noaa.gov/stp/solar/solarradio.html).

Daily observations are collected from 2419 surface observation stations in China to obtain the daily mean (Tave), maximum (Tmax), and minimum (Tmin) temperatures. These daily data are quality controlled and homogeneity adjusted at China’s National Meteorological Information Center (NMIC), with most station data available after the late 1950s (Cao et al., 2016). To investigate the influences of the SC and PVO on extreme temperatures in China, we also calculate four percentile-based extreme temperature indices recommended by tthe Joint WMO Commission for Climatology (CCL)–Climate Va-riability and Predictability (CLIVAR)–Joint Commission for Oceanography and Marine Meteorology (JCOMM) Expert Team on Climate Change Detection and Indices (ETCCDI). These extreme indices defined by ETCCDI describe different types of extreme events in the observations, including frequency, intensity, and duration, based on daily data (Zhang et al., 2011; Sillmann et al., 2013a, b; Donat et al., 2014). Here, percentile-based indices are used to describe the variations in the frequency of extreme temperatures. The four indices are calculated by counting the percentage (%) of days when daily mini-mum/maximum temperature is below the 10th percentile (Tn10p and Tx10p, cold nights and days) or above the 90th percentile (Tn90p and Tx90p, warm nights and days) of daily minimum/maximum temperatures during a year. The percentile thresholds are chosen according to the observations from the 1961–90 base period because data from almost all the stations are available in China for this period. These data are thus representative of the upper and lower tails of the distribution functions of daytime and nighttime temperatures, which have experienced significant shifts in most regions of the globe (Donat and Alexander, 2012). The percentile index calculations are conducted according to Zhang et al. (2011) to avoid inhomogeneity in the indices at the boundaries between the base and out-of-base periods. All the temperature indices are calculated for each station first, and then anomalies are obtained by removing the 1981–2000 long-term mean.

We employ EOF analysis in this study to extract the variation characteristics of the coupled stratosphere–troposphere system. EOF analyses are carried out on the continuous monthly 17-level geopotential height (GH) anomalies in the Northern Hemisphere during 1979–2015. In this way, we can obtain a set of time series for each EOF mode, which could represent the variation features of the stratosphere–troposphere system. Then, correlation and regression analyses are conducted based on these EOF time series to investigate the connections between the GH anomalies and the near-surface temperature observations, as well as the related physical mechanism. On the other hand, to identify the driving source and vertical structure of the second EOF mode in the GH anomalies of the coupled stratosphere–troposphere system, EOF analyses are conducted on the GH anomalies layer by layer from the stratosphere to lower levels over the same time period and horizontal domain as in the earlier analyses.

3 Significant variations of the stratosphere– troposphere system

The stratosphere and troposphere are regarded as a coupled system, and EOF analyses are conducted on the monthly 17-level GH anomalies during 1979–2015. Figure 1 shows distributions of the first and second EOF modes at different levels. These two modes account for 31.29% and 17.39% of the total variance, respectively. The distributions of the first mode in both the stratosphere and troposphere clearly present the NAM feature. Positive GH anomalies occupy the polar areas from near the ground (figure omitted) to the upper stratosphere (Figs. 1ac) and negative GH anomalies are located around the polar vortex in the midlatitudes, corresponding to the negative phase of the NAM. The first EOF mode and its corresponding time series mainly represent polar vortex variation features that were investigated in detail in our previous study (Lu et al., 2016).

The distribution of GH anomalies of the second EOF mode exhibits a multipole pattern in the Northern Hemisphere. To concentrate on the East Asian climate features, here only the Eurasian regions are displayed (Figs. 1df). Negative GH anomalies are located in the polar, subtropical, and equatorial regions, while positive GH anomalies are situated in the midlatitudes. Anomalous GH centers with opposite signs are alternatingly distributed from north to south, and the GH distributions display a barotropic feature in the middle–high latitudes. This distribution pattern is opposite to that of the first mode in middle–high latitudes, indicating that when the time series of the first mode and that of the second mode are in the same phase, their resulting circulation anomalies may offset each other. In contrast, when the time series of these two modes are in opposite phases, the resulting disturbances may be enhanced via superposition.

Figure 1 Distributions of the (a, b, c) first and (d, e, f) second EOF modes of the geopotential height (GH) anomalies (gpm) at (a, d) 50, (b, e) 100, and (c, f) 300 hPa in the Northern Hemisphere.

The time series of the first (Ha_1) and second (Ha_2) modes are compared with each other month by month during 1979–2015 in Fig. 2. These two time series have different active periods, during which their variation amplitudes become fairly large. The main active periods for Ha_1 occur in the late winter (red histogram in Fig. 2), meaning that most of the influences of the first EOF mode are concentrated from January to March. However, the strong variations in Ha_2 are mostly observed in the late autumn and early winter (October to December, green histogram in Fig. 2), indicating that the primary variation and effect of the second EOF mode occur during this period. These changing characteristics of Ha_1 and Ha_2 indicate that the first and second EOF modes of the GH anomalies both have important effects on the variations of the stratospheric and tropospheric circulations in winter and their influences are almost independent of each other. In addition, another notable feature of Ha_2 is that its time series exhibits an approximate decadal variation pattern, with larger positive values observed in the beginning periods of the 1980s, 1990s, and 2000s, and apparent negative values mainly in the late stages of each decade.

Figure 2 Time series of the first (red line) and second (green line) EOF modes of the GH anomalies during (a) 1979–90, (b) 1991–2000, and (c) 2001–15. Magenta histogram represents the value of Ha_1 in January–March, and green histogram indicates the value of Ha_2 in October–December. Both time series are normalized.

To further illustrate the periodicity of the second EOF mode, we carry out a Morlet wavelet analysis on Ha_2 (Fig. 3a). The areas shaded in red in Fig. 3a are the local wavelet power spectrum of Ha_2, indicating its significant periods and corresponding occurrence time. The regions within the black contours represent the values significant above the 95% confidence level. There appear two significant periods in Ha_2, both of which pass the statistical confidence test. One significant period is 60–80 months (approximately 5–7 yr) in 1984–90 and 2000–05. The other period, which is quite significant, is 118–132 months (approximately 10 yr) in 1995–2005, further demonstrating the decadal variation of the second EOF mode of the stratosphere–troposphere system.

Figure 3 (a) Morlet wavelet analysis results for the second EOF mode time series (areas within black contour lines indicate statistically significant anomalies at the 95% confidence level). (b) Temporal evolution of the 10.7-cm solar radiation flux. (c) The second EOF mode during 1979–2015. In (b, c), green and orange histograms represent the 10.7-cm solar radiation flux and Ha_2 in October–December, respectively; and green and red dashed lines indicate 120- to 144-month bandpass-filtered results. Both time series in (b) and (c) are normalized.

Figures 3b and 3c display distributions of the 10.7-cm solar radiation flux and Ha_2 from 1979 to 2015, respectively. The values of these two indices in October–December are shown in the form of histogram. The green and red dashed lines indicate the results after a 120- to 144-month bandpass filter. These two time series have similar features, with stronger solar activity and larger positive values of Ha_2 showing up mostly in the early stages of the 1980s, 1990s, 2000s, and 2010s. In the medium term, the 10.7-cm solar radiation flux and Ha_2 begin to significantly decline and exhibit large negative values in the later stages. Correlation analyses further prove the connection of these two time series, especially during October–December. Their correlation coefficient increases from 0.21 (using all monthly series) to 0.43 (using only the 3-month series), indicating that the decadal characteristic of Ha_2 is more significant in the late autumn and early winter. From the bandpass-filtered results, we can further see that the 10.7-cm solar radiation flux is slightly ahead of the time series of the second EOF mode of Ha_2, indicating that solar activity, as an important external forcing, plays a significant role in the distributions of the second EOF mode of the coupled stratosphere–troposphere system.

According to previous studies (Liu and Lu, 2010; Chen and Zhou, 2012; Kodera et al., 2016), the influences of the 11-yr SC usually affect the stratospheric circulations first and then gradually propagate down to the entire stratosphere–troposphere system. Therefore, we conduct EOF analysis of the GH anomalies for 1979–2015 on each vertical level, respectively. Figure 4 shows temporal evolutions of the second EOF mode at different levels, its respective variance contribution, and the correlation coefficient with Ha_2. The second EOF mode time series in the stratosphere (e.g., at 20 hPa) is very similar to Ha_2 and exhibits oscillations at a period of approximately 10 yr. Moreover, with decreasing altitude, this variation feature gradually weakens, and the correlation coefficient decreases. In the upper troposphere and lower stratosphere (UTLS, 200 hPa), the decadal variation of the second EOF time series becomes indistinct. This further indicates that the decadal variation characteristic of the second EOF mode of the coupled stratosphere–troposphere system is closely related to the 11-yr SC. The influences of solar activity are the most clear in the stratosphere and can be propagated downward with decreasing intensity.

Figure 4 Time series of the second EOF mode of the GH anomalies computed at different levels during 1979–2015. The black dashed lines indicate 120- to 144-month bandpass-filtered results. All time series are normalized.

According to the above analyses, the first (Ha_1) and second (Ha_2) EOF time series are referred to as the PVO index and the SCC index, respectively. These two indices have different active times: the main disturbances in the PVO index are concentrated in the period from January to March; while the primary activity of the SCC index occurs in late autumn and early winter (October–December). Will these properties propagate downward to the ground and affect the near-surface temperature? In the next section, we further examine the influences of the SCC index on the observed near-ground winter temperature in China and the mechanism behind this interaction.

4 Influences of the SCC and PVO indices

Figure 5 shows distributions of the correlation coefficients between the SCC index and the anomalies of Tave, Tmin, and Tmax in early winter during 1979–2015 as calculated from the observational data from 2419 surface meteorological stations. Here, we only display the results that exceed the 95% confidence level. Negative correlation coefficients with strong statistical reliability dominate the northern areas of China to the north of 35°N, and the ranges of the negative values are much broader in the plots of Tave and Tmin (Figs. 5a, b). In most areas, the correlation coefficients are larger than 0.3, reaching or exceeding –0.5over a number of stations. This result indicates that when the SCC index increases notably, the near-ground temperature in most parts of northern China will become lower than usual. The responses of the daily mean and minimum temperatures are more significant than that of the daily maximum temperature. Similar calculations based on the grid data from Wu and Gao (2013) were carried out to explore the relationship between the SCC index and observed temperatures, and consistent conclusions were obtained (figures omitted). These results further prove the cooling effects of strong solar activity on near-ground temperatures in early winter in China, especially in northern China.

Figure 5 Distributions of the correlation coefficients between the SCC index and Tave, Tmin, and Tmax anomalies from station observations in early winter (October–December) for the period 1979–2015. All indices are detrended before correlation analysis.

We also compute the correlation coefficient between the SCC index and the extreme temperature indices (Tn10p and Tx10p, cold nights and days, respectively; Tn90p and Tx90p, warm nights and days, respectively) in early winter. Figure 6 shows the results that pass the statistical test. The connections between the SCC index and warm nights/days are much closer. Negative correlation coefficients are observed in most regions of northern China, indicating that strong solar activity can reduce the frequency of warm nights and days in these areas in early winter. Positive correlation coefficients between the SCC index and cold nights/days can be seen in Figs. 6a, c. Significant positive values in Tn10p (Fig. 6a) are mainly located in the northern and eastern parts of China, while there are only a few stations for which the results pass the statistical test in Tx10p (Fig. 6c). This indicates that during active SC periods, cold nights will become more frequent in northern and eastern China, while cold days have much weaker responses.

Figure 6 Distributions of the correlation coefficients between the SCC index and extreme temperature indices derived from station observations in early winter (October–December) from 1979 to 2015: (a) Tn10p; (b) Tn90p; (c) Tx10p; and (d) Tx90p. All indices are detrended before correlation analysis.

The above correlation analyses of the near-ground temperature and extreme temperature indices indicate that the second EOF mode of the coupled stratosphere–troposphere system, which is mainly influenced by the 11-yr SC, corresponds to the cooling effects in China in early winter and that the major sensitive regions are concentrated in the northern parts of China. To determine the cause of this phenomenon, we present regressions of the early winter zonal wind anomalies against the second EOF time series in Fig. 7. The subtropical westerly jet in the UTLS zone is strong, and in the vertical direction, the zonal wind anomalies exhibit a barotropic feature with the same large-value, same-sign centers across different altitudes. At 200 hPa (Fig. 7b), the black dashed frame indicates a sensitive area reflecting the intensity of the westerly jet, and the two blue dashed frames denote the wind shear zone reflecting the position of the westerly jet, which can be referred to in Liang and Wang (1998) and Mao et al. (2007). The distributions of zonal wind anomalies in these areas indicate that when the SCC index increases notably (a period of strong solar activity), the subtropical westerly jet in East Asia will be stronger and move southward. The convergence of airflows at the entrance of the westerly jet can result in a clear increase in pressure at the lower levels, while the divergence at the jet exit can induce a decrease in pressure at the lower levels. Therefore, the enhanced pressure differences between land and ocean and the reinforcement of the East Asian trough jointly lead to a stronger winter monsoon, corresponding to lower near-ground temperatures from north to south in China.

Figure 7 Regressions of zonal wind (m s–1) anomalies against the second EOF time series at (a) 100, (b) 200, (c) 300, and (d) 500 hPa in early winter (October–December) during 1979–2015. Stippled regions indicate the regression coefficients statistically significant at the 95% confidence level, as estimated by the Student’s t-test. White contours represent the climatological westerly jet. See text for explanations to the black and blue boxes in (b).

The UV range of the solar radiation increases apparently in strong period of the SC (Chandra and McPeters, 1994) and this radiative energy is mainly absorbed by the stratospheric ozone layer. This extra-enhanced energy is helpful to the transformation of the stratospheric circulation from summer pattern to winter pattern (Andrews et al., 1987; Shindell et al., 2009). Month by month comparisons of the subtropical westerly jet (figure omitted) indicate that the jet intensities are usually larger in winter months than in autumn months. Accordingly, accelerated seasonal transformation of the stratosphere, induced by the stronger solar activity, corresponds to the enhanced subtropical westerly jet in the UTLS zone in late autumn and early winter.

To further demonstrate this deduction, we composite the zonal wind anomalies during October–December based on the normalized SCC index, and the differences between positive (SCC index > 1) and negative (SCC index < –1) years. Figure 8 shows the composite distribution of zonal wind anomalies at 200 hPa and the vertical structure of the zonal mean zonal wind anomalies. Both the horizontal and vertical distributions illustrate a reinforcement of the subtropical westerly jet in the active period of the SCC index, consistent with the above regression results. To summarize, strong solar activity can obviously change the stratospheric circulation, leading to enhancement of the westerly jet in early winter. This then strengthens the winter monsoon with enhanced cold air activities, and causes the cooling effects on the observed temperatures in early winter in northern China.

Figure 8 Composite distributions of (a) zonal wind anomalies (m s–1) at 200 hPa and (b) vertical structure of zonal mean zonal wind anomalies (m s–1) in early winter (October–December). The areas within white contour lines in (a) and the grey shaded parts in (b) indicate statistically significant anomalies at the 95% confidence level based on the Student’s t-test.

We also calculate the correlation coefficient between the PVO index and anomalies in the temperature and extreme temperature indices in late winter (January– March). Figure 9 shows the results that exceed the 95% confidence level. The influences of the PVO index on the near-ground temperature are mainly concentrated in the northeastern parts of China. For Tave, Tmin, and Tmax, significant negative correlation coefficients are observed in northeastern China, indicating that when the PVO index increases notably (corresponding to a weaker polar vortex with larger meridional extent), the daily mean, minimum, and maximum temperatures become lower than usual in Northeast China. Tx10p and Tn10p exhibit significant positive correlation coefficients in most parts of northeastern China, indicating that an enhanced polar vortex (decreased PVO index and reduced meridional extent) can reduce the frequency of cold days and nights in these areas. Most of the correlation coefficients between the PVO index and Tn90p and Tx90p fail the statistical test, meaning that the variations in the polar vortex do not significantly influence the frequency of warm nights and days. The above analyses illustrate that a weaker polar vortex corresponds to the cooling in Northeast China and vice versa. This is consistent with the conclusions of previous studies (Yin et al., 2013; Lu and Ding, 2015; Lu et al., 2016). Disturbances in the polar vortex usually begin in the stratosphere, and the GH anomalies can propagate downward in the troposphere and southward to the northeastern parts of the Eurasian continent, resulting in reinforcement of the East Asian trough and apparent cooling processes there.

Figure 9 Distributions of the correlation coefficients between the PVO index and temperature indices derived from station observations in late winter (January–March) during 1979–2015: (a) Tave; (b) Tmin; (c) Tmax; (d) Tx10p; and (e) Tn10p. All indices are detrended before correlation analysis.
5 Conclusions and discussion

In this paper, we analyzed the significant variations in the coupled stratosphere–troposphere system and diagnosed the influences of these variations on the near-ground temperature and extreme temperature indices in winter in China. Monthly NCEP-2 data and daily observations from 2419 surface meteorological observation stations in China are utilized in the analyses. Based on EOF analysis of the GH anomalies, the first EOF mode mainly represents the distribution and evolution characteristics of the polar vortex. The main active periods of the first EOF mode are concentrated in the late winter, from January to March, and disturbances in the polar vortex first occur in the stratosphere. Subsequently, circulation anomalies can propagate downward and southward, affecting the tropospheric midlatitudes. Correlation analyses assessing the relationship between the first EOF time series and the observed near-ground temperature indices in China indicate that a weak polar vortex process can lead to statistically significant cooling effects in northeastern China. The daily mean, minimum, and maximum temperatures all become lower than usual, and the cold days and nights become more frequent. A strong polar vortex confers the opposite effects.

The results of the Morlet wavelet analysis and sensitivity tests indicate that the second EOF mode of the coupled stratosphere–troposphere system is closely related to the 11-yr solar cycle and that the major active periods are concentrated in autumn and early winter from October to December. An increased amount of UV radiation, caused by stronger solar activity, will be absorbed by the stratospheric ozone layer, which helps to accelerate the transformation of the stratospheric circulation from summer pattern to winter pattern. This stratospheric variation corresponds to a stronger subtropical westerly jet in the UTLS zone in late autumn and early winter, bringing cooling effects on the near-ground temperature in China, especially in northern China. Correlation analyses show that during strong solar activity years, the near-ground temperature is lower than usual in most parts of China, especially the daily mean and minimum temperatures; simultaneously, the frequency of warm nights and days is significantly reduced, and cold nights and days are more frequent.

As a fundamental energy source, solar activity has great influences on the earth’s climate system through diverse approaches (Xiao et al., 2017). In terms of the 11-yr SC, the 10.7-cm solar radiation flux is commonly employed to investigate the “top-down” influence of solar UV irradiance that propagates from the stratosphere to troposphere (Ineson et al., 2011; Kodera et al., 2016). This mainly affects the large-scale circulation systems, such as planetary waves (Lu et al., 2017) and AO (Chen and Zhou, 2012). But it is not easy to find out direct connections between the 10.7-cm solar flux and near-ground weather or climate elements. The SCC index proposed in the current study is derived from the variation features of the coupled stratosphere–troposphere system. Although it is not as good as the 10.7-cm solar flux in describing the SC, it has a better ability to represent the influencing process of solar activity on near-ground temperature and extreme temperature events in boreal winter in China.

In view of the polar vortex influences, we also compared the PVO index presented in this study and the AO index as well as the NAM index. The PVO index has periods and regions of influence similar to those of the AO index, but the regions of influence of the AO index are a little broader, including not only the northeastern parts but also some northern and eastern regions of China. The NAM index, usually derived with multi-layer data, is better at expressing the propagation of the polar vortex disturbance from the stratosphere to troposphere, but when describing the effects on the near-ground systems/elements, the PVO index is more effective and has a better capability.

In this study, the first and second EOF modes of the GH anomalies have different active times and regions. Further analyses demonstrate that these two modes can represent the influences of the polar vortex and solar activity, respectively. Therefore, the time series of these two modes can be defined as indices for the PVO and SC activity, which can provide significant indications of the cooling process in different periods of boreal winter in China. Furthermore, in consideration that more comprehensive factors will provide a better picture of the variations in cooling processes in China, this line of research needs to be extended in the future.

Acknowledgments. The authors are grateful to the anonymous reviewers for constructive comments and to the Editor for helpful assistance. The meteorological analysis data used in this study were kindly provided by the NCAR/NCEP of USA and the NMIC of China.

References
Alexander, L. V., X. B. Zhang, T. C. Peterson, et al., 2006: Global observed changes in daily climate extremes of temperature and precipitation. J. Geophys. Res., 111, D05109. DOI:10.1029/2005JD006290
Andrews, D. G., J. R. Holton, and C. B. Leovy, 1987: Middle Atmosphere Dynamics. Academic Press, San Diego, 489 pp.
Baldwin, M. P., and T. J. Dunkerton, 1999: Propagation of the Arctic Oscillation from the stratosphere to the troposphere. J. Geophys. Res., 104, 30,937–30,946. DOI:10.1029/1999JD900445
Baldwin, M. P., D. B. Stephenson, D. W. J. Thompson, et al., 2003: Stratospheric memory and skill of extended-range weather forecasts. Science, 301, 636–640. DOI:10.1126/science.1087143
Brown, S. J., J. Caesar, and C. A. T. Ferro, 2008: Global changes in extreme daily temperature since 1950. J. Geophys. Res., 113, D05115. DOI:10.1029/2006JD008091
Caesar, J., L. Alexander, and R. Vose, 2006: Large-scale changes in observed daily maximum and minimum temperatures: Creation and analysis of a new gridded data set. J. Geophys. Res., 111, D05101. DOI:10.1029/2005JD006280
Cai, M., and R.-C. Ren, 2007: Meridional and downward propagation of atmospheric circulation anomalies. Part I: Northern Hemisphere cold season variability. J. Atmos. Sci., 64, 1880–1901. DOI:10.1175/JAS3922.1
Cao, L. J., Y. N. Zhu, G. L. Tang, et al., 2016: Climatic warming in China according to a homogenized data set from 2419 stations. Int. J. Climatol., 36, 4384–4392. DOI:10.1002/joc.4639
Chandra, S., and R. D. McPeters, 1994: The solar cycle variation of ozone in the stratosphere inferred from Nimbus 7 and NOAA 11 satellites. J. Geophys. Res., 99, 20,665–20,671. DOI:10.1029/94JD02010
Chen, W., and Q. Zhou, 2012: Modulation of the Arctic Oscillation and the East Asian winter climate relationships by the 11-yr solar cycle. Adv. Atmos. Sci., 29, 217–226. DOI:10.1007/s00376-011-1095-3
Chen, W., X. Q. Lan, L. Wang, et al., 2013: The combined effects of the ENSO and Arctic Oscillation on the winter climate anomalies in East Asia. Chinese Sci. Bull., 58, 1355–1362. DOI:10.1007/s11434-012-5654-5
Donat, M. G., and L. V. Alexander, 2012: The shifting probability distribution of global daytime and night-time temperatures. Geophys. Res. Lett., 39, L14707. DOI:10.1029/2012GL052459
Donat, M. G., J. Sillmann, S. Wild, et al., 2014: Consistency of temperature and precipitation extremes across various global gridded in situ and reanalysis datasets. J. Climate, 27, 5019–5035. DOI:10.1175/JCLI-D-13-00405.1
Gray, L. J., J. Beer, M. Geller, et al., 2010: Solar influences on climate. Rev. Geophys., 48, RG4001. DOI:10.1029/2009RG000282
Hartmann D. L., A. M. G. Klein Tank, M. Rusticucci, et al., 2013: Observations: Atmosphere and surface. Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. T. Stocker, D. Qin, G. Plattner, et al., Eds., Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 159–254, doi: 10.1017/CBO9781107415324.008.
Ineson, S., A. A. Scaife, J. R. Knight, et al., 2011: Solar forcing of winter climate variability in the Northern Hemisphere. Nat. Geosci., 4, 753–757. DOI:10.1038/ngeo1282
Jiang, Q., X. K. Ma, and F. Wang, 2016: Analysis of the January 2016 atmospheric circulation and weather. Meteor. Mon., 42, 514–520. DOI:10.7519/j.issn.1000-0526.2016.04.016
Kanamitsu, M., W. Ebisuzaki, J. Woollen, et al., 2002: NCEP–DOE AMIP-II reanalysis (R-2). Bull. Amer. Meteor. Soc., 83, 1631–1643. DOI:10.1175/BAMS-83-11-1631
Kodera, K, 2002: Solar cycle modulation of the North Atlantic Oscillation: Implication in the spatial structure of the NAO. Geophys. Res. Lett., 29, 1218. DOI:10.1029/2001GL014557
Kodera, K., 2003: Solar influence on the spatial structure of the NAO during the winter 1900–1999. Geophys. Res. Lett., 30, 1175. DOI:10.1029/2002GL016584
Kodera, K., and Y. Kuroda, 2002: Dynamical response to the solar cycle. J. Geophys. Res., 107, 4749. DOI:10.1029/2002JD002224
Kodera, K., R. Thiéblemont, S. Yukimoto, et al., 2016: How can we understand the global distribution of the solar cycle signal on the Earth’s surface?. Atmos. Chem. Phys., 16, 12,925–12,944. DOI:10.5194/acp-16-12925-2016
Labitzke, K., 1987: Sunspots, the QBO, and the stratospheric temperature in the North Polar Region. Geophys. Res. Lett., 14, 535–537. DOI:10.1029/GL014i005p00535
Labitzke, K., 2001: The global signal of the 11-year sunspot cycle in the stratosphere: Differences between solar maxima and minima. Meteor. Z., 10, 83–90. DOI:10.1127/0941-2948/2001/0010-0083
Labitzke, K., 2003: The global signal of the 11-year sunspot cycle in the atmosphere: When do we need the QBO. Meteor. Z., 12, 209–216. DOI:10.1127/0941-2948/2003/0012-0211
Labitzke, K., 2005: On the solar cycle–QBO relationship: A summary. J. Atmos. Sol.-Terr. Phys., 67, 45–54. DOI:10.1016/j.jastp.2004.07.016
Li, F., H. J. Wang, and Y. Q. Gao, 2014: On the strengthened relationship between the East Asian winter monsoon and Arctic Oscillation: A comparison of 1950–70 and 1983–2012. J. Climate, 27, 5075–5091. DOI:10.1175/JCLI-D-13-00335.1
Liang, X.-Z., and W.-C. Wang, 1998: Associations between China monsoon rainfall and tropospheric jets. Quart. J. Roy. Meteor. Soc., 124, 2597–2623. DOI:10.1002/qj.49712455204
Liu, Y., and C. H. Lu, 2010: The influence of the 11-year sunspot cycle on the atmospheric circulation during winter. Chinese J. Geophys., 53, 1269–1277. DOI:10.3969/j.issn.0001-5733.2010.06.004
Lu, C. H., and Y. H. Ding, 2015: Analysis of isentropic potential vorticities for the relationship between stratospheric anomalies and the cooling process in China. Sci. Bull., 60, 726–738. DOI:10.1007/s11434-015-0757-4
Lu, C. H., B. T. Zhou, and Y. H. Ding, 2016: Decadal variation of the Northern Hemisphere annular mode and its influence on the East Asian trough. J. Meteor. Res., 30, 584–597. DOI:10.1007/s13351-016-5105-3
Lu, H., L. J. Gray, and I. P. White, 2017: Stratospheric response to the 11-yr solar cycle: Breaking planetary waves, internal reflection, and resonance. J. Climate, 30, 7169–7190. DOI:10.1175/JCLI-D-17-0023.1
Mao, R., D. Y. Gong, and Q. M. Fang, 2007: Influences of the East Asian jet stream on winter climate in China. J. Appl. Meteor. Sci., 18, 137–146.
Matthes, K., B. Funke, M. E. Andersson, et al., 2017: Solar forcing for CMIP6 (v3.2). Geosci. Model Devel., 10, 2247–2302. DOI:10.5194/gmd-10-2247-2017
Park, H.-J., and J.-B. Ahn, 2016: Combined effect of the Arctic Oscillation and the western Pacific pattern on East Asia winter temperature. Climate Dyn., 46, 3205–3221. DOI:10.1007/s00382-015-2763-2
Park, T.-W., C.-H. Ho, and S. Yang, 2011: Relationship between the Arctic Oscillation and cold surges over East Asia. J. Climate, 24, 68–83. DOI:10.1175/2010JCLI3529.1
Shindell, D., D. Rind, N. Balachandran, et al., 2009: Solar cycle variability, ozone, and climate. Science, 284, 305–308. DOI:10.1126/science.284.5412.305
Sillmann, J., V. V. Kharin, F. W. Zwiers, et al., 2013a: Climate extremes indices in the CMIP5 multimodel ensemble: Part 2. Future climate projections. J. Geophys. Res., 118, 2473–2493. DOI:10.1002/jgrd.50188
Sillmann, J., V. V. Kharin, X. Zhang, et al., 2013b: Climate extremes indices in the CMIP5 multimodel ensemble: Part 1. Model evaluation in the present climate. J. Geophys. Res., 118, 1716–1733. DOI:10.1002/jgrd.50203
Sun, Y., T. Hu, X. B. Zhang, et al., 2018: Anthropogenic influence on the eastern China 2016 super cold surge. Bull. Amer. Meteor. Soc., 99, S123–S127. DOI:10.1175/BAMS-D-17-0092.1
Thompson, D. W. J., and J. M. Wallace, 1998: The Arctic Oscillation signature in the wintertime geopotential height and temperature fields. Geophys. Res. Lett., 25, 1297–1300. DOI:10.1029/98GL00950
Wang, R. L., Z. N. Xiao, K. Y. Zhu, et al., 2015: Asymmetric impact of solar activity on the East Asian winter climate and its possible mechanism. Chinese J. Atmos. Sci., 39, 815–826. DOI:10.3878/j.issn.1006-9895.1410.14211
Wu, J., and X. J. Gao, 2013: A gridded daily observation dataset over China and comparison with the other datasets. Chinese J. Geophys., 56, 1102–1111. DOI:10.6038/cjg20130406
Xiao, Z.-N., D.-L. Li, L.-M. Zhou, et al., 2017: Interdisciplinary studies of solar activity and climate change. Atmos. Ocean. Sci. Lett., 104, 325–328. DOI:10.1080/16742834.2017.1321951
Yin, S., J. Feng, and J. P. Li, 2013: Influences of the preceding winter Northern Hemisphere annular mode on the spring extreme low temperature events in the north of eastern China. Acta Meteor. Sinica, 71, 96–108. DOI:10.11676/qxxb2013.008
Zhang, X. B., L. Alexander, G. C. Hegerl, et al., 2011: Indices for monitoring changes in extremes based on daily temperature and precipitation data. Wiley Interdisciplinary Reviews: Climate Change, 2, 851–870. DOI:10.1002/wcc.147
Zhao, L., J. S. Wang, H. W. Liu, et al., 2017: Amplification of the solar signal in the summer monsoon rainband in China by synergistic actions of different dynamical responses. J. Meteor. Res., 31, 61–72. DOI:10.1007/s13351-016-6046-6
Zhou, B. Z., L. H. Gu, Y. H. Ding, et al., 2011: The great 2008 Chinese ice storm: Its socioeconomic–ecological impact and sustainability lessons learned. Bull. Amer. Meteor. Soc., 92, 47–60. DOI:10.1175/2010BAMS2857.1