1. Introduction
The possible preservation of solar cycle signals in precipitation has been reported from several monsoon regions, including South Asia, West Africa, and North America (e.g., Kodera 2004; van Loon et al. 2004; Kodera and Shibata 2006; Narasimha and Bhattacharyya 2010; van Loon and Meehl 2012). This resembles the relationship between the solar cycle and other climate phenomena; however, the apparent correlation between the solar cycle and monsoon precipitation may differ over time and from region to region, even to the extent of adjacent regions having correlations with the opposite sign (e.g., Thresher 2002; Meehl et al. 2009). Zhao et al. (2012) reported that the correlation between the sunspot number (SSN) and precipitation in June shows regional differences among the boundary, interior of the East Asian summer monsoon (EASM), and nonmonsoon regions of China. Wang and Zhao (2012) confirmed these differences and also the occurrence of latitudinal and longitudinal variations in the relationship between the solar signal and the monsoon and atmospheric circulation over the sunspot cycle. They suggested that in high SSN years (HSYs), the domain area of the EASM is likely to be larger and located farther north, and that the western Pacific subtropical high (WPSH) is also likely to be larger than during low SSN years (LSYs). Moreover, a stronger monsoon flow over the Bay of Bengal region and the larger WPSH in the HSYs probably contribute to the northward expansion of the June EASM during the HSYs.
Mei-yu is a unique rainy season within the seasonal northward march of the EASM, often making the most persistent rain events over East Asia (Ding and Chan 2005; Chen and Chang 1980). Mei-yu is also the most important rainy season for the East Asian monsoonal region because the mei-yu precipitation usually contributes to the local precipitation maximum each year in East Asia (Chen and Chang 1980), as the Indian summer monsoon precipitation does in Indian subcontinent. The beginning of the mei-yu season signifies the onset of the northern summer monsoon in East Asia (e.g., Chen and Chang 1980). Generally, after the onset in the South China Sea in mid-May, the EASM propagates northward and then begins to affect mainland China; subsequently, a large-scale quasi-zonal monsoon rainband is sequentially established over south China and Taiwan, the Yangtze and Huaihe River basins and Japan, and the Korean Peninsula (Ding and Chan 2005). Different terminologies are used for the same meteorological phenomenon over different geographical regions (Chen 1994; Ding et al. 2007), such as mei-yu in China, baiu in Japan, and changma in Korea (Oh et al. 1997; Qian and Lee 2000; Chen 2004; Ninomiya 2004; Ding et al. 2007). For simplicity, these monsoon rain belts are all referred to here as the East Asian mei-yu rainband. It is characterized by a large-scale quasi-zonal monsoon rainband with a quasi-stationary front and cloud zone (Fig. 1b), which is distinct from other rainy seasons (Lau and Li 1984; Wang and LinHo 2002; Ding and Chan 2005; Ding et al. 2007). The cloud zone in Fig. 1b, matching the mei-yu rainband in Fig. 1a, represents the mei-yu front that usually remains over East Asia and moves slowly northward after it establishes. These features are similar to those of the Indian maximum cloud band that propagates northward on a 30–40-day cycle during the Indian monsoon (Sikka and Gadgil 1980). It is presumable that the mei-yu rainband during the typical rainy season necessarily involves important information of the EASM variation. Thus, a study of the variability of the East Asian mei-yu rainband could improve our understanding of the nature and extent of solar cycle imprints on the EASM and be very useful for climate prediction and disaster prevention.
(a) Distribution of observatories in the study area (boxed area: 20°–45°N, 105°–122.5°E) and precipitation on 5 Jul 2013 using the China NMIC merged precipitation dataset of Chinese autoweather station precipitation and CMORPH precipitation product. (b) Chinese FY2E satellite infrared image on 1300 UTC 5 Jul 2013.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
(a) Distribution of observatories in the study area (boxed area: 20°–45°N, 105°–122.5°E) and precipitation on 5 Jul 2013 using the China NMIC merged precipitation dataset of Chinese autoweather station precipitation and CMORPH precipitation product. (b) Chinese FY2E satellite infrared image on 1300 UTC 5 Jul 2013.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
(a) Distribution of observatories in the study area (boxed area: 20°–45°N, 105°–122.5°E) and precipitation on 5 Jul 2013 using the China NMIC merged precipitation dataset of Chinese autoweather station precipitation and CMORPH precipitation product. (b) Chinese FY2E satellite infrared image on 1300 UTC 5 Jul 2013.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
In this paper, we analyze daily precipitation in China over the period 1958–2012 to explore the possible connections between the East Asian mei-yu rainband in middle–eastern China (major monsoon region) during the summer half-year (1 April to 30 September) and the SSN.
2. Data and method
We used the yearly mean relative SSN as the primary proxy of solar activity, and this was obtained from the U.S. National Oceanic and Atmospheric Administration (NOAA) National Geophysical Data Center (NGDC) (
The precipitation data in Fig. 1a on 5 July 2013 are derived from the China NMIC merged precipitation dataset of Chinese auto–weather station precipitation and the U.S. Climate Prediction Center morphing technique (CMORPH) precipitation product (Pan et al. 2012), which includes precipitation over the East Asian continent and ocean since 1 January 2008. An infrared image obtained from the Chinese FY2E satellite on 1300 UTC 5 July 2013 is also used in Fig. 1b.
Daily meridional wind velocity υ (17 pressure levels) and the pressure coordinate vertical velocity ω (12 pressure levels) at 2.5° longitude by 2.5° latitude grid points since 1958 from the National Centers for Environmental Prediction (NCEP)–National Center for Atmospheric Research (NCAR) reanalysis dataset (Kalnay et al. 1996) are also used in this study.
The study area (20°–45°N, 105°–122.5°E) was selected as it contains the main land area affected by the mei-yu rainband (outlined by the solid line in Fig. 1a). The precipitation data from the stations were interpolated onto a 0.5° × 0.5° grid using the Cressman interpolation technique (Cressman 1959).
The mean latitude of the rainband (MLRB) was obtained from the interpolated grid precipitation data. For each year, we selected each date span from within the summer half-year (1 April to 30 September), and the MLRB during this span was defined as the mean latitude of maximum precipitation at each longitude (with an interval of ±0.5°) in the area enclosed by the box in Fig. 1a. The start and end date of the span can be any day between 1 April and 30 September (183 days in total), so the shortest time span would be 1 day and the longest would be 183 days. For example, the number of all the date spans that start on 1 April is 183 in total (the shortest date span is 1 April, the longest is from 1 April to 30 September), and the number of those that start on 2 April is 182 in total. So for each year the total number of all possible date spans between 1 April and 30 September is 16 836 (=183 + 182 + 181 + … +1).
MLRBs with each the same span (e.g., starts on 18 April and ends on 26 June), but from all different years (1958–2012), form an MLRB time series with 55 [=(2012 − 1958) + 1] elements, so a total of 16 836 series were obtained. Then, the correlation coefficient between each series and the SSN from 1958 to 2012 was calculated. In total, 16 836 correlation coefficients are obtained.
3. Major results
The final results are shown as color shading in Fig. 2. The x and y axes indicate the start and end dates, respectively, of the particular time spans. There are two distinct high-correlation areas corresponding to spans that both start around late May, but end in mid-July and late August. Among all the 16 836 correlation coefficients, the maximum correlation coefficient of 0.47 (>99.9% confidence level) is associated with the span between 22 May and 13 July (53 days).
Correlations between the annual SSN and yearly MLRB series for any time span in the northern summer half-year (1 Apr to 30 Sep). The shortest duration of the span is 1 day, and the longest span is 183 days, in each year from 1958 to 2012. The x and y axes indicate the start and end dates of the span, respectively. The correlation coefficients of the 95%, 99%, and 99.9% confidence levels for the 55 yr are 0.26, 0.34, and 0.43, respectively. The maximum correlation coefficient of 0.47 (arrow) corresponds to the span from 22 May to 13 Jul.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
Correlations between the annual SSN and yearly MLRB series for any time span in the northern summer half-year (1 Apr to 30 Sep). The shortest duration of the span is 1 day, and the longest span is 183 days, in each year from 1958 to 2012. The x and y axes indicate the start and end dates of the span, respectively. The correlation coefficients of the 95%, 99%, and 99.9% confidence levels for the 55 yr are 0.26, 0.34, and 0.43, respectively. The maximum correlation coefficient of 0.47 (arrow) corresponds to the span from 22 May to 13 Jul.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
Correlations between the annual SSN and yearly MLRB series for any time span in the northern summer half-year (1 Apr to 30 Sep). The shortest duration of the span is 1 day, and the longest span is 183 days, in each year from 1958 to 2012. The x and y axes indicate the start and end dates of the span, respectively. The correlation coefficients of the 95%, 99%, and 99.9% confidence levels for the 55 yr are 0.26, 0.34, and 0.43, respectively. The maximum correlation coefficient of 0.47 (arrow) corresponds to the span from 22 May to 13 Jul.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
It is interesting that the start date of the highest-correlation span, 22 May, is also the average onset date of the mei-yu in south China (i.e., the average start date of the period when the EASM and a large-scale quasi-zonal monsoon rainband influence mainland China) (e.g., Tao and Chen 1987; Lau and Yang 1997; Wang and LinHo 2002; Chen 2004; Zheng et al. 2006), and the end date, 13 July, is also the average date when the mei-yu rains ended in mainland China during the period 1958–2000 (Xu et al. 2001). That is to say, the span with the highest correlation corresponds to a special climatological period: the East Asian mei-yu season. This strongly suggests that the high correlation during the mei-yu season has an explicit physical significance rather than just a statistical significance.
After the mei-yu season, the EASM advances rapidly and jumps northward into North China. It remains there for around 1 month, but no stable large-scale quasi-zonal rainband develops such as that seen during the mei-yu season. A break in the monsoonal rainy period occurs from late July to early August in some East Asian monsoon regions (Chen 2004). Correspondingly, the MLRB exhibits a rather weak correlation with the SSN, as shown in Fig. 2. When the EASM retreats toward south China in mid- to late August and reestablishes a monsoon rainband there (Ding and Chan 2005), the high correlation between MLRB and SSN is also reestablished, as is indicated by the second high-correlation region in Fig. 2.
The date span with the highest-correlation coefficient (0.47) between the MLRB and SSN (i.e., the East Asian mei-yu season, 22 May to 13 July) was the focus of further analysis. The mei-yu season MLRB in each year and the annual SSN are compared from 1958 onward in Fig. 3. One can note that the mei-yu MLRB is often more to the north around the SSN peaks than that around the SSN valleys. However, examining the two time series, it almost looks as if there is a different regime beginning around the mid-1990s, as multiple peaks appear thereafter, even around the SSN valleys (e.g., 1995 and 1996), which leads to a relatively high-frequency oscillation. We find that the correlations for the period of 1958–94 and 1995–2012 are 0.54 and 0.33, respectively. The weaker correlation for 1995–2012 than that for 1958–94 resulted from the relatively high-frequency interannual oscillation of the raw data, which should not be related to the 11-yr sunspot cycle but, however, could be related to El Niño–Southern Oscillation (ENSO) and the quasi-biennial oscillation (QBO).
Time series of annual SSN (shaded), unfiltered East Asian mei-yu season MLRB (dotted curve with circles), and 8-yr low-pass filtered East Asian mei-yu season MLRB (thick solid curve) with its mean values (thick horizontal dashed lines) in the HSYs and LSYs of each solar cycle between 1958 and 2012. The thin solid horizontal line denotes the mean SSN (67.5) and the mean mei-yu MLRB (26.7°N) for 1958–2012. The East Asian mei-yu season is the span from 22 May to 13 Jul.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
Time series of annual SSN (shaded), unfiltered East Asian mei-yu season MLRB (dotted curve with circles), and 8-yr low-pass filtered East Asian mei-yu season MLRB (thick solid curve) with its mean values (thick horizontal dashed lines) in the HSYs and LSYs of each solar cycle between 1958 and 2012. The thin solid horizontal line denotes the mean SSN (67.5) and the mean mei-yu MLRB (26.7°N) for 1958–2012. The East Asian mei-yu season is the span from 22 May to 13 Jul.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
Time series of annual SSN (shaded), unfiltered East Asian mei-yu season MLRB (dotted curve with circles), and 8-yr low-pass filtered East Asian mei-yu season MLRB (thick solid curve) with its mean values (thick horizontal dashed lines) in the HSYs and LSYs of each solar cycle between 1958 and 2012. The thin solid horizontal line denotes the mean SSN (67.5) and the mean mei-yu MLRB (26.7°N) for 1958–2012. The East Asian mei-yu season is the span from 22 May to 13 Jul.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
To identify the deduction, the global wavelet spectra of unfiltered annual SSN and the mei-yu MLRB during 1958–2012 using Morlet’s wavelet as the mother wavelet are shown in Fig. 4 (cf. Torrence and Compo 1998). Both of the highest spectrum peaks of the SSN and the mei-yu MLRB time series are about 11 yr at >95% and about 82% red noise confidence levels, respectively, although the confidence level of the latter is not very high. Besides, 2–5-yr spectra of the mei-yu MLRB are at >95% red noise confidence level (95% confidence-level line for MLRB not shown), which is likely related to ENSO and QBO. So to remove the possible effects of ENSO and the QBO, an 8-yr low-pass filtered MLRB series was derived and is also shown in Fig. 3. The correlation coefficient between the unfiltered SSN and the 8-yr low-pass filtered MLRB is 0.87 for the entire period of 1958–2012. It indicates that the low-frequency components of the MLRB are more strongly correlated with the SSN than the unfiltered MLRB. As the confidence intervals of correlations were different for the unfiltered and filtered data, Monte Carlo tests were used to define their confidence intervals (cf. Zhao et al. 2012). The Monte Carlo tests were calibrated against noise (spectrally and amplitude-wise matched), because when sample sizes are not very large a simple Monte Carlo test can be misleading (cf. Bhattacharyya and Narasimha 2007). The threshold values of the correlation coefficients at the 99.9% confidence level were 0.43 and 0.74 for the unfiltered and filtered data, respectively, for the entire period of 1958–2012. Therefore, the correlation coefficients for the unfiltered and 8-yr low-pass filtered MLRB are both highly statistically significant, which indicates that there are significant solar cycle imprints in the mei-yu rainbands. Moreover, the correlations between the filtered MLRB series and the SSN for the period of 1958–94 and 1995–2012 are 0.89 and 0.84, respectively. Such substantial enhancements of correlation using filtered data, especially for the period of 1995–2012 (from 0.33 to 0.84), also suggest stronger solar signals in the MLRB on decadal and interdecadal time scales than the interannual time scale.
Global wavelet spectra of unfiltered annual SSN (thick solid lines) and the unfiltered mei-yu MLRB (thick dashed lines; mm2) during 1958–2012 using Morlet’s wavelet as the mother wavelet.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
Global wavelet spectra of unfiltered annual SSN (thick solid lines) and the unfiltered mei-yu MLRB (thick dashed lines; mm2) during 1958–2012 using Morlet’s wavelet as the mother wavelet.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
Global wavelet spectra of unfiltered annual SSN (thick solid lines) and the unfiltered mei-yu MLRB (thick dashed lines; mm2) during 1958–2012 using Morlet’s wavelet as the mother wavelet.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
The difference in behavior of the mei-yu MLRB during the HSYs and LSYs was also tested. The mean mei-yu MLRBs for the 23 samples from the years with an annual SSN greater than the mean SSN and for the 32 samples from the years with an annual SSN less than the mean SSN during the period 1958–2012 were 27.4° and 26.2°N, respectively. This indicates that the mei-yu MLRBs in HSYs are located about 1.2° farther north than in LSYs. A Student’s t test, which is particularly suitable for testing the statistical significance of the difference between the mean values of two series, was applied to the two sample groups and indicated that the latitude difference is statistically significant at a confidence level of 99%. The meridional shift of 1.2° (about 130 km) of the rainband is important information for climate prediction and flood prevention. Because the position of the mei-yu rainband is always a key problem for rainy season prediction each year, which largely determines the precipitation spatial distribution pattern during the rainy season. Moreover, regional heavy rain events and floods during the East Asian monsoon are frequently closely related to the position and duration of the mei-yu rainband (Si et al. 2009).
In addition, the amplitude of the interannual variation of the mei-yu MLRB in the HSYs was larger than in the LSYs. The mei-yu MLRB variances during the HSYs and LSYs were 2.46 and 0.77, respectively, and the F statistic, which tests the statistical significance of the difference between the variances of two series, indicated that the amplitude of the interannual variation of the mei-yu MLRB in the HSYs was significantly larger than in the LSYs (at a confidence level of 99%). The two characteristics together suggest that the probability that the MLRB is farther south in the LSYs than usual is larger than that the MLRB is farther north in the HSYs. This is also useful information for practical applications.
4. Verification
To verify the relationship between the mei-yu rainband and solar activity and diagnose the related spatial structure, the correlation coefficient between the MLRB and precipitation of China (Fig. 5), υ (Fig. 6), or ω (Fig. 8) at each grid point during the East Asian mei-yu season is calculated.
Correlation coefficient between (a) unfiltered MLRB or (b) annual SSN and the precipitation of China at each grid point during the East Asian mei-yu season for 1958–2012. Absolute values less than 0.2 are omitted and the interval of contour is 0.2. Negative values are indicated by dashed lines. Lighter and darker red (positive correlation) or blue (negative correlation) shaded areas indicate regions where the correlation is significant at the 90% and 95% confidence levels, respectively. The long dashed lines indicate the East Asian mei-yu MLRB averaged over 1958–2012.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
Correlation coefficient between (a) unfiltered MLRB or (b) annual SSN and the precipitation of China at each grid point during the East Asian mei-yu season for 1958–2012. Absolute values less than 0.2 are omitted and the interval of contour is 0.2. Negative values are indicated by dashed lines. Lighter and darker red (positive correlation) or blue (negative correlation) shaded areas indicate regions where the correlation is significant at the 90% and 95% confidence levels, respectively. The long dashed lines indicate the East Asian mei-yu MLRB averaged over 1958–2012.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
Correlation coefficient between (a) unfiltered MLRB or (b) annual SSN and the precipitation of China at each grid point during the East Asian mei-yu season for 1958–2012. Absolute values less than 0.2 are omitted and the interval of contour is 0.2. Negative values are indicated by dashed lines. Lighter and darker red (positive correlation) or blue (negative correlation) shaded areas indicate regions where the correlation is significant at the 90% and 95% confidence levels, respectively. The long dashed lines indicate the East Asian mei-yu MLRB averaged over 1958–2012.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
As in Fig. 5, but for the correlations with υ at (a),(b) 1000 hPa and in the (c),(d) vertical cross section along 105°–122.5°E.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
As in Fig. 5, but for the correlations with υ at (a),(b) 1000 hPa and in the (c),(d) vertical cross section along 105°–122.5°E.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
As in Fig. 5, but for the correlations with υ at (a),(b) 1000 hPa and in the (c),(d) vertical cross section along 105°–122.5°E.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
The correlation map of precipitation (Fig. 5a) shows a seesaw between the north (positive correlation) and south (negative correlation) of the 55-yr mean mei-yu MLRB for 1958–2012 (long-dashed line). Figure 5b shows that the correlation pattern is basically consistent with that of SSN, although the latter is not as strong as the former. Moreover, it is also similar to the result of Wang and Zhao (2012) with June precipitation data of 1901–2006, suggesting the relationship between solar activity and precipitation should be stable.
Figure 6a shows a positive correlation (northward wind) zone near the mean mei-yu MLRB and a strong negative correlation (southward wind) zone to the south of it over the South China Sea at 1000 hPa. The correlation relationship is highlighted by Fig. 6b with the SSN. The distribution is inclined to rain more heavily (lightly) near and to the north of the mean mei-yu MLRB (in south China and the South China Sea) during the HSYs than during the LSYs, causing the mei-yu rainband to shift poleward (equatorward). The correlations in the vertical cross section along 105°–122.5°E (Figs. 6c,d) further this relation. Moreover, they both show that significant negative correlations (southward wind) in the lower troposphere and significant positive correlations (northward wind) near the 20-hPa stratosphere over south China constitute an up–down seesaw. Although the up–down seesaw does not denote that the upper-northward wind and the lower-southward wind definitely form a closed meridional cell between them because the level of 20 hPa is too high, the revealed pattern is consistent with the Hadley cell and the difference cell in Wang and Zhao (2012) and is the reverse relative to the normal monsoonal cell.
It is a matter worthy of note that although Fig. 6b shows that there is a close relationship between the cross-equatorial Somali jet and the SSN that had been validated, which probably links with solar activity (Kodera 2004), the relationship is not found in Fig. 6a. This suggests that although both the mei-yu MLRB and the Somali jet are highly correlated with the SSN, there is not a significant teleconnection between them. This is interesting, but not necessarily surprising as solar activity may affect different regional circulations that may be independent of each other, unless such regional circulations are affected by an overarching global-scale circulation. For example, in the Indian monsoons it has been suggested that the displacement of the Hadley cell could explain the kind of regional differences found in the effect of solar activity on subcontinental rainfall (Bhattacharyya and Narasimha 2005; Haigh et al. 2005), similar to those indicated in the present paper. However, both Figs. 6c and 6d show a common up–down seesaw between the stratosphere at about 25°N and low troposphere at about 20°N over the EASM region, which suggests the meridional wind in the two zones is connected with not only the monsoon rainband but also solar activity. So we define an index for the up–down seesaw pattern (SI) as the difference of υ in the vertical cross section along 105°–122.5°E between the point (25°N, 20 hPa) and the point (20°N, 1000 hPa) during the East Asian mei-yu season. The correlations at the two, that is, upper and lower, grid points with the SSN (the mei-yu MLRB) are 0.28 and −0.33 (0.33 and −0.29), respectively. They are all at about the 97% confidence level. Figure 7 shows the series of the SI, the mei-yu MLRB, and the SSN from 1958 to 2012. The correlations of the SI with the SSN and the mei-yu MLRB are 0.42 (confidence level 99.8%) and 0.41 (confidence level 99.8%), respectively, which are both apparently higher than those at the two independent grid points. It signifies that the seesaw effect of υ between the stratosphere and troposphere is related to the SSN cycle; on the other hand, the meridional oscillation of the monsoon rainband is also related to the seesaw effect. More importantly, it implies that the seesaw is inclined to affect the monsoonal rainband as a whole because it statistically amplifies or strengthens both the regional circulation response to solar forcing and the monsoonal rainband response to the regional circulation, likely by the top–down mechanism and the bottom–up mechanism proposed by Meehl et al. (2009). Besides, the SI series since 1995 do not have obvious multiple peaks as the mei-yu MLRB, which likely hints that the vertical seesaw could affect the mei-yu rainband on decadal time scales because the SI involves high-level signals that may not be very easily affected by ENSO. The same correlations with ω (Fig. 8) are calculated. Figure 8a with ω at 500 hPa and Fig. 8c with ω in the vertical cross section along 105°–122.5°E from 1000 to 100 hPa confirm the north–south seesaw of Fig. 5 and show that monsoon convective activity is weak over south China as anomalous downward velocity (positive correlation) and strong to the north of the mean mei-yu MLRB (long-dashed lines) as anomalous upward velocity (negative correlation). The same correlations with the SSN (Figs. 8b,d) further confirm the relationship and show some other strong correlation zones, for example, south of the Somali Peninsula in the equatorial Southern Hemisphere. The pattern of ω over the EASM region is also inclined to cause the meridional oscillation of the mei-yu rainband. The weak monsoon convective activity over south China during the East Asian mei-yu season of the HSYs is likely related to anomalous northward wind in the stratosphere, which is opposite the direction of the climatological near-tropopause returning monsoon flow shown in Wang and Zhao (2012). Besides, the results of the correlation analysis of precipitation, υ, and ω for the period 1958–94 and 1995–2012 (figures not shown) are all similar to those for the entire period 1958–2012, especially for 1958–94, although the unfiltered mei-yu MLRB series seems to have a different regime since 1995, as seen in Fig. 3.
Time series of annual SSN (shaded), unfiltered East Asian mei-yu MLRB (dotted curve with circles), and SI (thick solid curve) between 1958 and 2012.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
Time series of annual SSN (shaded), unfiltered East Asian mei-yu MLRB (dotted curve with circles), and SI (thick solid curve) between 1958 and 2012.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
Time series of annual SSN (shaded), unfiltered East Asian mei-yu MLRB (dotted curve with circles), and SI (thick solid curve) between 1958 and 2012.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
As in Fig. 5, but for the correlations with ω at (a),(b) 500 hPa and in the (c),(d) vertical cross section along 105°–122.5°E. Positive correlation denotes anomalous downward velocity.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
As in Fig. 5, but for the correlations with ω at (a),(b) 500 hPa and in the (c),(d) vertical cross section along 105°–122.5°E. Positive correlation denotes anomalous downward velocity.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
As in Fig. 5, but for the correlations with ω at (a),(b) 500 hPa and in the (c),(d) vertical cross section along 105°–122.5°E. Positive correlation denotes anomalous downward velocity.
Citation: Journal of Climate 27, 8; 10.1175/JCLI-D-13-00482.1
5. Discussion
The behavior of the climatic systems in the troposphere is, of course, dominated by the driving dynamic processes, while external forcing can only exert its influence through coupling with these dominant factor(s). Compared with the dominant factors, the solar cycle forcing is distant and weak and hence is still poorly understood. Based on an analysis of the regional differences in the correlation between June precipitation in China and the solar cycle, Zhao et al. (2012) and Wang and Zhao (2012) suggested that the boundary regions of different climate systems seem to be more sensitive to solar cycle effects because the dominant factors are much weaker there than inside the systems. However, potential solar cycle imprints could be easily obscured by the natural variability of the systems. Nevertheless, if these boundaries move significantly with the solar cycle, their behavioral differences could reveal the solar cycle imprints. Many studies present evidence of a poleward shift or expansion in the high solar activity conditions of the Indian monsoon (Kodera 2004; van Loon and Meehl 2012), the Hadley cell, the subtropical jet, the Ferrel cell (Haigh 2003; Gleisner and Thejll 2003; Haigh et al. 2005; Brönnimann et al. 2007), and so on, and an equatorward shift in the low solar activity conditions of North Atlantic storm tracks (Martin-Puertas et al. 2012). The differences in the location of the mei-yu rainband under different solar conditions presented in this paper are consistent with the oscillation of the northern boundary of the EASM related to the solar cycle as reported by Zhao et al. (2012) and Wang and Zhao (2012). However, the issue of why some climate systems move during different phases of the solar cycle remains the key question and requires further investigation.
6. Summary
In this paper, the latitude of the monsoon rainband of different time spans during the summer half-year from 1958 to 2012 in China was compared with the SSN. A significant relation was found between the MLRB and the SSN when a prevailing large-scale quasi-zonal rainband develops, that is, the East Asian mei-yu season (22 May to 13 July) and also during the period when the EASM retreats to south China in mid- to late August and reestablishes a monsoon rainband there. The maximum correlation occurred during the East Asian mei-yu season. The mei-yu rainband moves northward and has a larger interannual variability during HSYs than during LSYs. The two high-correlation periods correspond to the two periods in which large-scale monsoon rainbands prevail and not with the entire EASM period. This correspondence between the statistical and physical evidence implies that solar activity dominates the decadal variability of the EASM, especially the mei-yu regime, but not during the non-mei-yu period of the summer half-year. The meridional and vertical wind analysis further verifies the strong response of the monsoon rainband and suggests that the forcing signal is likely related to an up–down seesaw and a north–south seesaw.
Acknowledgments
We thank the NMIC of China for the precipitation dataset and NGDC for the SSN data and NCEP/NCAR for wind data. We thank the two anonymous reviewers and the editor for providing very useful critical comments and constructive suggestions on the paper. Wavelet analysis software was provided by C. Torrence and G. Compo (http://paos.colorado.edu/research/wavelets/). This research was supported by the National Natural Science Foundation of China (41305131 and 40931056), the National Basic Research Program of China [2012CB957800 (801, 803, 804) and 2012CB417205], and the China Special Fund for Meteorological Research in the Public Interest (GYHY201406020).
REFERENCES
Bhattacharyya, S., and R. Narasimha, 2005: Possible association between Indian monsoon rainfall and solar activity. Geophys. Res. Lett., 32, L05813, doi:10.1029/2004GL021044.
Bhattacharyya, S., and R. Narasimha, 2007: Regional differentiation in multidecadal connections between Indian monsoon rainfall and solar activity. J. Geophys. Res., 112, D24103, doi:10.1029/2006JD008353.
Brönnimann, S., T. Ewen, T. Griesser, and R. Jenne, 2007: Multidecadal signal of solar variability in the upper troposphere during the 20th century. Space Sci. Rev., 125, 305–317, doi:10.1007/s11214-006-9065-2.
Chen, G. T. J., 1994: Large-scale circulations associated with the East Asian summer monsoon and the mei-yu over south China and Taiwan. J. Meteor. Soc. Japan, 72, 959–983.
Chen, G. T. J., 2004: Research on the phenomena of meiyu during the past quarter century: An overview. East Asian Monsoon, C. P. Chang, Ed., World Scientific Publishing, 404–435.
Chen, G. T. J., and C. P. Chang, 1980: The structure and vorticity budget of an early summer monsoon trough (mei-yu) over southeastern China and Japan. Mon. Wea. Rev., 108, 942–953, doi:10.1175/1520-0493(1980)108<0942:TSAVBO>2.0.CO;2.
Cressman, G. P., 1959: An operational objective analysis system. Mon. Wea. Rev., 87, 367–374, doi:10.1175/1520-0493(1959)087<0367:AOOAS>2.0.CO;2.
Ding, Y. H., and J. C. L. Chan, 2005: The East Asian summer monsoon: An overview. Meteor. Atmos. Phys., 89, 117–142, doi:10.1007/s00703-005-0125-z.
Ding, Y. H., J. J. Liu, Y. Sun, Y. J. Liu, J. H. He, and Y. F. Song, 2007: A study of the synoptic-climatology of the meiyu system in East Asia (in Chinese). Chin. J. Atmos. Sci., 31, 1082–1101.
Gleisner, H., and P. Thejll, 2003: Patterns of tropospheric response to solar variability. Geophys. Res. Lett., 30, 1711, doi:10.1029/2003GL017129.
Haigh, J. D., 2003: The effects of solar variability on the earth’s climate. Philos. Trans. Roy. Soc. London, 361, 95–111, doi:10.1098/rsta.2002.1111.
Haigh, J. D., M. Blackburn, and R. Day, 2005: The response of tropospheric circulation to perturbations in lower-stratospheric temperature. J. Climate, 18, 3672–3685, doi:10.1175/JCLI3472.1.
Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77, 437–471, doi:10.1175/1520-0477(1996)077<0437:TNYRP>2.0.CO;2.
Kodera, K., 2004: Solar influence on the Indian Ocean monsoon through dynamical processes. Geophys. Res. Lett., 31, L24209, doi:10.1029/2004GL020928.
Kodera, K., and K. Shibata, 2006: Solar influence on the tropical stratosphere and troposphere in the northern summer. Geophys. Res. Lett., 33, L19704, doi:10.1029/2006GL026659.
Lau, K. M., and M. T. Li, 1984: The monsoon of East Asia and its global association. Bull. Amer. Meteor. Soc., 65, 114–125, doi:10.1175/1520-0477(1984)065<0114:TMOEAA>2.0.CO;2.
Lau, K. M., and S. Yang, 1997: Climatology and interannual variability of the southeast Asian summer monsoon. Adv. Atmos. Sci., 14, 141–162, doi:10.1007/s00376-997-0016-y.
Martin-Puertas, C., and Coauthors, 2012: Regional atmospheric circulation shifts induced by a grand solar minimum. Nat. Geosci., 5, 397–401, doi:10.1038/ngeo1460.
Meehl, G. A., J. M. Arblaster, K. Matthes, F. Sassi, and H. Van Loon, 2009: Amplifying the Pacific climate system response to a small 11-year solar cycle forcing. Science, 325, 1114–1118, doi:10.1126/science.1172872.
Narasimha, R., and S. Bhattacharyya, 2010: A wavelet cross-spectral analysis of solar–ENSO–rainfall connections in the Indian monsoons. Appl. Comput. Harmonic Anal., 28, 285–295, doi:10.1016/j.acha.2010.02.005.
Ninomiya, K., 2004: Large- and mesoscale features of meiyu/baiu front associated with intense rainfalls. East Asian Monsoon, C. P. Chang, Ed., World Scientific Publishing, 404–435.
Oh, J. H., W. T. Kwon, and S. B. Ryoo, 1997: Review of the researches on changma and future observational study (KORMEX). Adv. Atmos. Sci., 14, 207–222, doi:10.1007/s00376-997-0020-2.
Pan, Y., Y. Shen, J. J. Yu, and P. Zhao, 2012: Analysis of the combined gauge-satellite hourly precipitation over China based on the OI technique. Acta Meteor. Sin., 70, 1381–1389.
Qian, W. H., and D. K. Lee, 2000: Seasonal march of Asian summer monsoon. Int. J. Climatol., 20, 1371–1378, doi:10.1002/1097-0088(200009)20:11<1371::AID-JOC538>3.0.CO;2-V.
Si, D., Y. H. Ding, and Y. J. Liu, 2009: Decadal northward shift of the meiyu belt and the possible cause (in Chinese). Chin. Sci. Bull., 54, 4742–4748, doi:10.1007/s11434-009-0385-y.
Sikka, D. R., and S. Gadgil, 1980: On the maximum cloud zone and the ITCZ over Indian longitudes during the southwest monsoon. Mon. Wea. Rev., 108, 1840–1853, doi:10.1175/1520-0493(1980)108<1840:OTMCZA>2.0.CO;2.
Tao, S. Y., and L. X. Chen, 1987: A review of recent research on the East Asian summer monsoon in China. Monsoon Meteorology, C. P. Chang and T. N. Krishnamurti, Eds., Oxford University Press, 60–92.
Thresher, R. E., 2002: Solar correlates of southern hemisphere mid-latitude climate variability. Int. J. Climatol., 22, 901–915, doi:10.1002/joc.768.
Torrence, C., and G. P. Compo, 1998: A practical guide to wavelet analysis. Bull. Amer. Meteor. Soc., 79, 61–78, doi:10.1175/1520-0477(1998)079<0061:APGTWA>2.0.CO;2.
van Loon, H., and G. A. Meehl, 2012: The Indian summer monsoon during peaks in the 11 year sunspot cycle. Geophys. Res. Lett., 39, L13701, doi:10.1029/2012GL051977.
van Loon, H., G. A. Meehl, and J. M. Arblaster, 2004: A decadal solar effect in the tropics in July–August. J. Atmos. Sol. Terr. Phys., 66, 1767–1778, doi:10.1016/j.jastp.2004.06.003.
Wang, B., and LinHo, 2002: Rainy season of the Asian–Pacific summer monsoon. J. Climate, 15, 386–398, doi:10.1175/1520-0442(2002)015<0386:RSOTAP>2.0.CO;2.
Wang, J. S., and L. Zhao, 2012: Statistical tests for a correlation between decadal variation in June precipitation in China and sunspot number. J. Geophys. Res., 117, D23117, doi:10.1029/2012JD018074.
Xu, Q., Y. W. Yang, and Q. M. Yang, 2001: The meiyu in middle-lower reaches of Yangtze river during 116 recent years (I) (in Chinese). Torrential Disaster, 1, 44–53.
Zhao, L., J. S. Wang, and H. J. Zhao, 2012: Signature of the solar cycle on decadal variability in monsoon precipitation over China. J. Meteor. Soc. Japan, 90, 1–9, doi:10.2151/jmsj.2012-101.
Zheng, B., J. Y. Liang, A. L. Lin, C. H. Li, and D. J. Gu, 2006: Frontal rain and summer monsoon rain during pre-rainy season in south China. Part I: Determination of the division dates (in Chinese). Chin. J. Atmos. Sci., 30, 1207–1216.
