1. Introduction
Northeast China (NEC) is the major corn field of China. The variability in corn production, which affects the total agricultural production of the country significantly, is influenced strongly by the conditions of local temperature and precipitation. Because of the mutual influences by tropical and extratropical atmospheric systems, the climate over NEC is characterized by many complex features, and improved understanding of these features is important not only for climate prediction operations but also for studies of hydrology, ecology, and agriculture.
Previous studies have revealed many features of variations of NEC climate in station observations, reanalysis data, and numerical models. Many studies have focused on the synoptic and seasonal-to-interannual features of the regional climate and their relations with tropical and high-latitude influences (e.g., Lian and An 1998; Liu and Wang 2001). Recent investigations have also been conducted to understand the decadal changes, regime shift, and long-term trends of changes in temperature and precipitation (Zhang and Zhang 2005; Sun et al. 2006a,b; Yang et al. 2008). It is found that in the past century, there exists a positive trend in temperature, especially in winter, and a negative trend in total precipitation, especially in fall. However, as the regional climate becomes warmer, extreme precipitation events exhibit an increasing tendency.
Previous studies have also analyzed the variations of atmospheric circulation systems and their relationships with regional temperature and precipitation (e.g., Lian and An 1998; Sun et al. 2002). These studies attempt not only to understand the physical mechanisms responsible for variations of the regional climate but also to improve the prediction skill of the climate. Li et al. (2005) examined the relationships of NEC precipitation with El Niño–Southern Oscillation (ENSO), the North Pacific Oscillation (NPO), the Arctic Oscillation, the North Atlantic Oscillation, and the Pacific decadal oscillation. They found that NPO was correlated with the NEC precipitation most significantly, although multiple factors should be considered for predicting the regional climate. Overall, the effect of ENSO on extratropical NEC climate is weak, and the skill of predicting the regional climate by numerical models is usually low (Wang et al. 2008) because of various reasons: model deficiencies, strong irregular features of the interannual signals and thus low predictability, and others. Nevertheless, further investigations into the characteristics of NEC climate and its relationship with large-scale climate control are important for understanding the mechanisms responsible for the variability of the regional climate, improving the prediction of the climate, and mitigating the economic loss due to meteorological disasters.
In this study, we conduct an analysis of the detailed features of NEC climate variations in both time and frequency domains. We attempt to understand how the dominant climate features on different time scales and during different periods of time behave and are accompanied by different atmospheric circulation patterns. For example, even though the NEC climate is closely related to NPO, as seen in Li et al. (2005), does this relationship vary with time and change with time scales? It is important to answer the question because the various features of different time scales are often attributed by different causes. It is expected that the study will also provide useful information for improving the prediction of the regional climate over northeast China.
More specifically, we apply several advanced analysis tools to investigate the variations of temperature and precipitation over the city of Changchun and their relationships with large-scale circulation and sea surface temperature (SST) patterns. We reveal the characteristics of temperature and precipitation variations in time and frequency domains, focusing on the signals of dominant time scales, and assess the relationships of the dominant climate signals with large-scale circulation and SST patterns. For example, we explore the time–frequency characteristics of relationships between NPO and temperature/precipitation, which have not been explored before.
Since this study focuses on the detailed features of the variations of regional climate represented by time series of temperature, precipitation, and others, we mainly analyze the data over Changchun, instead of the area averages over all of NEC, for several reasons. First, values averaged over large spatial domain may smooth out some detailed features that would be otherwise revealed by individual station data. Second, Changchun is one of the few meteorological stations in NEC that have long records of meteorological observations. Third, the precipitation over NEC is often characterized by relatively homogeneous features in signifying droughts and floods, and the mean temperature over Changchun, located near the center of the Songliao Plains, represents well the warm and cold conditions over most of NEC (Ding 1980; Sun et al. 2000; Sun and Wang 2006; also see the discussion in section 2).
The rest of this paper is organized as follows. In section 2, we describe the basic features of datasets and analysis methods. In section 3, we depict the detailed features, including the time–frequency characteristics of Changchun temperature and precipitation. We discuss the relationships of the temperature and precipitation with local and large-scale circulation and SST features in section 4 and with NPO in section 5. A summary of the results obtained is provided in section 6.
2. Data and analysis methods
The main data used in this study are the daily precipitation and surface air temperature over Changchun of NEC. These data cover the period since 1909, with missing records during 1943–48. The data of Changchun station also include daily surface winds and sea level pressure over Changchun since 1980. The temperature and precipitation over 79 NEC stations are also used in this study. Among the 79 stations, there is only one station with a location change in 1957. Quality control of the datasets because of a systematic change in instrumentation and others has been conducted to meet the requirement of meteorological operations by the China Meteorological Administration.
Figure 1 shows the patterns of correlation of seasonal surface temperature between Changchun and other NEC stations for spring (March–May), summer (June–August), autumn (September–November), and winter (December–February). It can be seen that the variability of temperature over Changchun is significantly related to the variations of temperature over all of NEC year round. The correlation patterns show uniform features over a large domain, with the strongest relationship over the Songliao Plains (around 42.4°–48.7°N, 118.7°–128°E). More specifically, the coefficients of correlation between Changchun temperature and NEC-averaged temperature for 1961–2002 are 0.94, 0.98, 0.96, and 0.98 for winter, spring, summer, and autumn, respectively. There is also a strong relationship in seasonal precipitation (Fig. 2), especially in winter. As for temperature, the strongest relationship is found in the correlation between Changchun precipitation and the precipitation over the Songliao Plains. The coefficients of correlation between Changchun precipitation and NEC-averaged precipitation for 1961–2002 are 0.67, 0.65, 0.68, and 0.72 for winter, spring, summer, and autumn, respectively, and all these numbers significantly exceed the 99% confidence level. However, compared to the correlation of temperature, the correlation of precipitation is weaker and exhibits more spatially inhomogeneous features. In particular, a weak relationship appears between Changchun precipitation and the precipitation over northern NEC in summer (Fig. 2b).
The datasets analyzed in this study also include the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis (Kalnay et al. 1996) and the National Oceanic and Atmospheric Administration (NOAA) Extended Reconstructed Sea Surface Temperature (Smith and Reynolds 2003). The NPO index applied in the analysis is defined as the normalized difference in sea level pressure between 25° and 40°N, 130° and 170°E and 50° and 65°N, 130° and 170°E (also see Wallace and Gutzler 1981).
The study employs a number of statistical tools, which include wavelet analysis, least squares analysis, and methods of correlation, coherence, and regression analyses. Many features of these tools are discussed appropriately in the following sections [see Yang et al. (2007) for more a detailed description].
3. Time–frequency characteristics of temperature and precipitation variations
Figure 3 shows the climatological means of monthly temperature and precipitation over Changchun. For 1949–2006, January is the coldest (−15.6°C) and driest (0.14 mm day−1) month and July is the hottest (23.1°C) and wettest (5.56 mm day−1) month of the year. Extremely high temperature was observed in July 1999 (25.3°C) and extremely low temperature in January 1970 (−20.4°C). Extremely heavy precipitation occurred in July 1986, which is 14.3 mm day−1. For 1909–42, January (July) is also the coldest (hottest) and driest (wettest) month of the year. The temperature and precipitation are −16.8°C and 0.18 mm day−1 for January and 23.4°C and 5.60 mm day−1 for July, respectively. Figure 4 further shows the distributions of probability density function of temperature and precipitation over Changchun. In January, the local temperature is mostly in the range of −16.8° to −14.4°C, followed by a range of −19.2° to −16.8°C (Fig. 4a). In July, it is mostly in the range of 21.7°–22.7°C, followed by a range of 22.7°–23.7°C (Fig. 4b). For 1909–42, the temperature has the largest density in the range of −19.2° to −16.8°C for January and the range of 22.7°–23.7°C for July. For both 1909–42 and 1949–2006, the density of precipitation over Changchun is the largest in the range of 0–0.1 mm day−1 in January and 5.3–7.8 mm day−1 (followed by 2.7–5.3 mm day−1) for July (Figs. 4c,d). It should be pointed out that the data of 1909–42 may be less reliable compared to those of 1949–2006 and thus only the latter period is mainly analyzed in this study.
We apply wavelet transform (Morlet et al. 1982) and other related analyses to reveal the time–frequency characteristics of the variations of temperature and precipitation. Since a distorted edge effect of wavelet spectrum may occur in the transform, especially in the lower frequency bands, as in Yang et al. (2007), who analyzed the time–frequency characteristics of the U.S. Great Plains precipitation, we apply the leap-step time series analysis (LSTSA) model developed by Zheng et al. (2000) to improve the data information of end points. The LSTSA is a nonlinear model. It decomposes a time series into deterministic and stochastic components, and the stochastic component is further characterized by several stochastic models, each of which is valid within a subdomain of the time series.
Figure 5 shows the estimated time–frequency spectrum from the wavelet transform for temperature. Unsurprisingly, the variability of temperature is absolutely dominated by the annual cycle. Signals on intraseasonal, interannual, and interdecadal time scales can also be observed, despite their smaller mean power. The “first guess” provided by Fig. 5, largely qualitative, offers great efficiency to our next analysis in which the least squares method of Householder transforms (Powell and Reid 1969; also see Yang et al. 2007), a linear regression problem, to quantitatively determine the mean magnitude and phase of the strong oscillating signals of temperature variations. We focus on various time scales (see rows 3–7 in Table 1): semiannual, annual, quasibiennial, and interannual (about 4 and 7.5 yr). Since the periods of interannual oscillations are relatively unstable and drifts in frequency may occur in the spectral estimate, we determine their mean values by a method of trial and error in the process of least squares computations. We identify and determine the optimal mean periods of spectral signals by adjusting the periodic values step by step. In the process, the amplitudes and phases of the spectral signals are also estimated and the uncertainties in the period of the signals are measured by the standard deviations of phases.
As seen in Table 1, the mean amplitude is 19.1°C for the annual variation and 1.9°C for the semiannual signal. These two time scales have the most stable phase (see the smallest standard deviation of 0.01 in column 6). The table also shows clear signals on quasibiennial and interannual time scales. Row 9 indicates that, on average, these oscillatory signals explain 87% of the total variation of the temperature, which is obtained as the ratio of the mean temperature (13.65°C) minus root-mean-square (RMS; 1.76°C) to the mean. The RMS is calculated from the residual series after the five oscillating signals, and the constant and linear trend terms are removed from the original temperature data. In addition, there exists a positive trend in the temperature, which is 0.26°C decade−1 in 1909–42 but increases to 0.34°C decade−1 in 1949–2006 (see column 9). The results discussed earlier are for 1949–2006; however, similar features can also be found for 1909–42.
We also apply the same analysis to the precipitation over Changchun. Figure 6 shows that the variability of precipitation is also characterized by a strong annual cycle, although it is not as dominant as the annual cycle of temperature. As in the temperature field, signals of precipitation variations on intraseasonal, interannual, and interdecadal time scales, especially on the intraseasonal time scale, are also apparent. The mean amplitude of the semiannual signal (1 mm day−1) is about a half of that of the annual variation (2.13 mm day−1), as shown in Table 2. These two time scales also have the most stable phase. Compared to temperature, the biennial oscillation of precipitation has a shorter time scale of 1.53 yr. Overall, the oscillatory signals on semiannual, annual, biennial, and interannual time scales explain 40% of the total variations of precipitation. Table 2 also shows a small negative trend in the precipitation, which decreases at a rate of −0.03 mm day−1 decade−1 compared to −0.08 mm day−1 decade−1 in 1909–42.
4. Relationships with local and large-scale circulation features
a. Relationships with local winds
A similar analysis is also applied to precipitation. Figures 8a,b show that precipitation over Changchun increases when easterly and southerly winds prevail and decreases under the influences of westerly and northerly winds. The meridional wind is more significantly correlated with precipitation than with temperature. It leads to change in precipitation, indicating the importance of water vapor supply for precipitation over NEC. Like temperature, precipitation has significant relationships with the zonal and meridional winds on both annual and semiannual time scales (Figs. 8c,d). The strongest relationship is found on the annual time scale for precipitation and zonal wind but on the semiannual time scale for precipitation and meridional wind. Nevertheless, the relationships of winds with temperature and precipitation become weaker after the annual and semiannual time scales are removed.
b. Relationships with large-scale atmospheric and SST patterns
Then, we analyze the relationships of regional temperature and precipitation with 850-mb circulation and SST patterns. Figure 9a shows that, on the annual time scale, an increase in temperature is related to an anticyclonic pattern over the extratropical Pacific and a cyclonic pattern over Asia and the northern Indian Ocean. Changchun is located between these two circulation patterns. An increase in temperature is also related to warming (cooling) over the northern (southern) hemisphere and strong westerly flow over southern Asia, indicating a strong summer monsoon circulation. The figure also implies that in winter, the strong East Asian monsoon decreases the temperature over northeast China (Wang and Chen 2010).
On the semiannual time scale (Fig. 9b), the temperature increases when Changchun is controlled by an anticyclonic pattern whose center is over the warm waters northeast of Japan. In the meantime, a cyclonic pattern appears over Southeast Asia and the subtropical western Pacific, centered over 20°N, 135°E and accompanied by low SST. Figure 9b shows that westerly flow weakens over southern Asia and the northern Indian Ocean when the temperature increases over Changchun, suggesting the importance of extratropical atmospheric processes for affecting the variations of NEC temperature on semiannual time scale.
It can be seen from Fig. 10 that on the annual time scale, the large-scale circulation patterns associated with an increase in precipitation (Fig. 10a) are similar to those associated with an increase in temperature (Fig. 9a). An increase in precipitation is linked to the intensification of southeasterly flow related to the anomalous anticyclonic over the extratropical North Pacific and to the anomalous cyclonic pattern over the Asian continent and the southwesterly flow over southern Asia and the northern Indian Ocean. An increase in precipitation is also linked to the apparent warming in the subtropical–extratropical western Pacific.
On the semiannual time scale, an increase in precipitation is more directly associated with the water vapor supply from the Sea of Japan and from the east (Fig. 10b). This moisture transportation is linked to an anomalous cyclonic pattern over the extratropical North Pacific, where SST decreases. Figure 10b also suggests that the intensification of the large-scale Asian summer monsoon circulation increases the precipitation over NEC. Note also the apparent cooling in the Indian Ocean, driven by the strong monsoon flow.
5. Relationships with NPO
Figure 11 shows the time series of NPO and its estimated time–frequency spectrum from wavelet transform. The variations of NPO at various frequencies are mostly unstable with time. For example, the spectrum of the annual time scale before the mid-1960s is larger than that afterward. The largest spectrum appears on the time scales of months. Least squares analysis indicates that, on average, the amplitude of NPO is largest and most stable on the semiannual time scale (see Table 3), followed by the annual time scale. The amplitude of the annual time scale is about a half of the amplitude of the semiannual time scale. A 9-yr oscillation of NPO is also found. On this decadal time scale, the NPO is strongly related to the Pacific decadal oscillation.
Figure 12 shows the cross correlation and cross coherence between the Changchun temperature and precipitation and the NPO index. In the computations, the constant and linear trend terms have been removed as required by the statistical analysis methods. The mean annual cycles have also been removed from the original data to focus on the features of other time scales. There is a highly strong relationship between the Changchun temperature and NPO. The temperature increases when NPO is positive and the correlation between the two time series significantly exceeds the 99% confidence level (Fig. 12a). Clearly, the largest coherence appears in the semiannual frequency band (Fig. 12c). There is also a significant relationship between the Changchun precipitation and NPO (Fig. 12b), although it is weaker than the temperature–NPO relationship. The correlation between the two, which indicates that precipitation decreases when NPO is positive, also significantly exceeds the 99% confidence level. The strongest precipitation–NPO coherence appears in the frequency band of 5–7 months, the semiannual time scale (Fig. 12d).
It is known that the cross-coherence estimations shown in Fig. 12 only reveal the mean features of the entire time span, and that the estimations are unable to depict the stability and variability of these features in specific frequency bands with respect to the time process over the data span. Following Yang et al. (2007), we further explore the coherence features between NPO and Changchun temperature/precipitation in both time and frequency domains by applying a technique in which the coherence in time–frequency domains is derived by first estimating the values of a subseries of 10 yr and then successively moving the data points for subsequent estimations. Figure 13a shows that significant coherence between NPO and Changchun temperature appears mainly in the higher frequency bands. In particular, large coherence estimates appear in the frequency bands up to 7 months. They are unstable (varying with time) and have been large since the 1980s but small before that. This analysis thus manifests the advantage of the method in depicting the detailed features of the relationship between Changchun temperature and NPO.
A similar analysis for NPO and Changchun precipitation also reveals a strong relationship in the high-frequency hands and large (small) coherence since (before) the 1980s (Fig. 13b). The largest coherence appears in 1981–92, mainly in the frequency band of 6–9 months. In addition, the large coherence is also seen on interannual time scales, including the biennial time scale from the late 1960s to the early 1970s.
We further depict the large-scale atmospheric circulation and SST patterns associated with the temporally varying feature of the large coherence signals shown in Fig. 13. It can be seen from Fig. 14, which shows the mean patterns of 850-mb winds and SST for 1968–79 and their differences between 1981 and 1992 and 1968 and 1979, that atmospheric circulation and SST patterns exhibit many different features between the two periods [the selection of the two periods also addresses the potential problem of the NCEP–NCAR reanalysis, over Asia and prior to 1968, as demonstrated by Yang et al. (2002)]. As shown in the mean patterns for 1981–92, NEC, including Changchun, is clearly affected by the East Asian trough (Fig. 14a). The East Asian monsoon circulation also influences the NEC climate. In the difference field (Fig. 14b), a cyclonic pattern appears over NEC, suggesting that the trough was stronger during 1981–92 than during 1968–79. Given the significant relationship between NPO and NEC climate and recalling the strong semiannual signals in temperature, precipitation, NPO, and their interrelationships, a strong semiannual component of the variation of the East Asian trough can be expected. The anomalous flow over East Asia indicates the stronger East Asian winter monsoon or the weaker summer monsoon in 1981–92. It can also be seen from Fig. 14b that the easterly trade wind over the tropical Pacific and the westerlies over the Indian Ocean and tropical Asia was weaker in 1981–92, associated with a general increase in SST. However, no apparent feature can be found in the change of SST over the extratropical northwestern Pacific Ocean.
6. Summary and further discussion
In this study, we have applied several advanced analysis tools to depict the time–frequency characteristics of the variations of temperature and precipitation over the city of Changchun in northeast China and their associations with large-scale atmospheric and oceanic conditions. The analysis methods include wavelet transformation, least squares analysis, coherence analysis, and regression analysis. A multistage filter and a leap-step time series analysis technique have also been applied.
One of the most prominent features of the variations of Changchun temperature and precipitation is the strong semiannual signals. For example, the amplitude of the semiannual signal of precipitation is about half of that of the annual cycle. The relationships of Changchun temperature and precipitation with local winds and with large-scale atmospheric circulation and SST patterns on the semiannual time scale are also strong, only second to the strong relationships on the annual time scale.
On the annual time scale, the NEC climate is affected by both the thermal contrast between the Asian continent and tropical Indo-Pacific Oceans and the difference in thermal condition between the continent and the extratropical North Pacific. Increases in temperature and precipitation are associated with cyclonic and anticyclonic patterns over the Asian continent and the North Pacific, respectively, and with strong southerly flow over East Asia and the northwestern Pacific. Clearly, a decrease (increase) in winter temperature is linked to the intensification (reduction) of the East Asian winter monsoon, and a decrease (increase) in summer precipitation is related to the weakening (strengthening) of the Asian summer monsoon circulation. On the semiannual time scale, the variability of NEC climate is mainly affected by the large-scale circulation pattern centered over the North Pacific, with the western portion over northeast China, North and South Korea, and Japan. Under the influence of the cyclonic (anticyclonic) pattern, precipitation over Changchun increases (decreases) but the regional temperature decreases (increases). While temperature may be mainly affected by the extratropical process, the variability of NEC precipitation is also related to the tropical Asian monsoon circulation on the semiannual time scale. On both annual and semiannual time scales, increases in temperature and precipitation are associated with warming in the subtropical–extratropical northwestern Pacific Ocean.
Strong relationships occur between Changchun temperature/precipitation and NPO, especially in the frequency band up to 7 months. Temperature increases and precipitation decreases when NPO is positive. These relationships were weak before 1980; however, they become stronger afterward, associated with the strengthening of the East Asian trough.
It should be pointed out that our analysis is mainly for Changchun. Although we have shown that the temperature and precipitation are highly correlated with those over much of northeast China (see Figs. 1, 2), we also carry out an analysis of the area-averaged temperature and precipitation over NEC (79 stations). The variations of Changchun temperature and precipitation are similar to those of NEC-averaged temperature and precipitation, consistent with the features shown in Figs. 1, 2. For example, the variations of NEC temperature and precipitation are also characterized by the two strongest and most stable signals on annual and semiannual time scales. In particular, the amplitude of NEC temperature signal is 18.97° and 1.87°C for the annual and semiannual time scales, respectively, comparable with 19.06° and 1.86°C for the Changchun temperature. Similarly, the amplitude of NEC precipitation signal is 2.09 and 0.98 mm day−1 for the annual and semiannual time scales, comparable with 2.13 and 1.00 mm day−1 for the Changchun counterpart. Figure 15 presents the cross correlation and cross coherence between the temperature and precipitation over NEC (79 stations) and the NPO index. It shows features similar to those presented in Fig. 12 for Changchun temperature and precipitation. Namely, there are highly strong relationships between NEC temperature and NPO (Fig. 15a) and between NEC precipitation and NPO (Fig. 15b). Temperature increases and precipitation decreases when NPO is positive. As shown in Fig. 12, the largest coherences also appear in the semiannual frequency bands (Figs. 15c,d), although significant coherences can also be found on intraseasonal time scales.
However, even though strong signals of semiannual time scale appear from the variations of NEC temperature/precipitation and NPO and the relationships between them, we should not claim that NPO exerts an influence on the regional climate at the particular time scale. It is possible that the variations of both the regional climate and NPO are associated with the variability of other climate patterns on the preferred time scale. The physical mechanisms responsible for these semiannual signals and the application of these signals in the prediction of the regional climate deserve further investigations. It should also be pointed out that in this study, we have depicted the features of NEC climate and its relationship with NPO on a wide range of time scales with a focus on the semiannual time scale, instead of the interannual time scale that has been studied extensively. The features shown in this study should vary from season to season. For example, we have repeated the analysis of Fig. 12 using the seasonally stratified data for winter (December–February), spring (March–May), summer (June–August), and fall (September–November) and found that NPO is positively and significantly correlated with Changchun temperature in winter and summer. The correlation between NPO and Changchun seasonal-mean precipitation and the correlation between NPO and Changchun temperature in other seasons are insignificant. Nevertheless, the application of seasonally stratified data to some of the analyses, such as the technique shown in Fig. 13, requires much longer data records.
Acknowledgments
We are thankful for the helpful reviews by editor James Renwick and four anonymous reviewers. This study was partially supported by the Governor’s Foundation of Jilin province, China.
REFERENCES
Ding, S., 1980: The climatic analysis of low temperature in summer over the Northeast China and influence for agricultural product (in Chinese). Acta Meteor. Sin., 38 , 234–242.
Ding, X., D. Zheng, and S. Yang, 2002: Variations of the surface temperature in Hong Kong during the last century. Int. J. Climatol., 22 , 715–730.
Jenkins, G. M., and D. G. Watts, 1968: Spectral Analysis and Its Applications. Holden-Day, 525 pp.
Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437–471.
Li, Q., S. Yang, V. E. Kousky, R. W. Higgins, K-M. Lau, and P. Xie, 2005: Features of cross-Pacific climate shown in the variability of China and US precipitation. Int. J. Climatol., 25 , 1675–1696.
Lian, Y., and G. An, 1998: The relationship among East Asian summer monsoon, El Niño and low temperature in Songliao Plains Northeast China (in Chinese). Acta Meteor. Sin., 56 , 724–735.
Liu, S., and N. Wang, 2001: The impacts of antecedent ENSO event on air temperature over Northeast China in summer (in Chinese). J. Trop. Meteor., 17 , 314–319.
Morlet, J., G. Arens, E. Fourgeau, and D. Glard, 1982: Wave propagation and sampling theory—Part I: Complex signal and scattering in multilayered media. Geophysics, 47 , 203–221. doi:10.1190/1.1441328.
Powell, M. J. D., and J. K. Reid, 1969: On applying Householder transformations to linear least squares problems. Mathematics, Software, A. J. H. Morrell, Ed., Vol. 1, Information Processing 68: Proceedings of IFIP Congress 1968, North-Holland Publishing Company, 122–126.
Smith, T. M., and R. W. Reynolds, 2003: Extended reconstruction of global sea surface temperatures based on COADS data (1854–1997). J. Climate, 16 , 1495–1510.
Sun, F., J. Yuan, and S. Lu, 2006a: The change and test of climate in Northeast China over the last 100 years (in Chinese). Climatic Environ. Res., 11 , 101–108.
Sun, F., X. Yang, S. Lu, and S. Yang, 2006b: The contrast analysis on the average and extreme temperature trend in Northeast China (in Chinese). J. Sci. Meteor. Sin., 26 , 157–163.
Sun, J-Q., and H-J. Wang, 2006: Regional difference of summer air temperature anomalies in Northeast China and its relationship to atmospheric general circulation and sea surface temperature (in Chinese). Chin. J. Geophys., 49 , 662–671.
Sun, L., G. An, Y. Lian, B. Shen, and X. Tang, 2000: A study of the persistent activity of northeast cold vortex in summer and its general circulation anomaly characteristics (in Chinese). Acta Meteor. Sin., 55 , 70–82.
Sun, L., and Coauthors, 2002: The unusual characteristics of general circulation in drought and flooding years of Northeast China (in Chinese). Climatic Environ. Res., 7 , 102–113.
Thomason, D. J., 1982: Spectrum estimation and harmonic analysis. Proc. IEEE, 70 , 1055–1096.
Vondrák, J., 1977: Problem of smoothing observational data II. Bull. Astron. Inst. Czech., 28 , 83–93.
Wallace, J. M., and D. S. Gutzler, 1981: Teleconnections in the geopotential field height during the Northern Hemisphere winter. Mon. Wea. Rev., 109 , 784–812.
Wang, H., J. Sun, X. Lang, L. Chen, and W. Fu, 2008: Some new results in the research of the international climate variability and short-term climate prediction (in Chinese). Chin. J. Atmos. Sci., 32 , 806–814.
Wang, L., and W. Chen, 2010: How well do existing indices measure the strength of the East Asian winter monsoon? Adv. Atmos., 27 , 855–875.
Yang, S., K-M. Lau, and K-M. Kim, 2002: Variations of the East Asian jet stream and Asian–Pacific–American winter climate anomalies. J. Climate, 15 , 306–325.
Yang, S., X. Ding, D. Zheng, and Q. Li, 2007: Depiction of the variations of Great Plains precipitation and its relationship with tropical central-eastern Pacific SST. J. Appl. Meteor. Climatol., 46 , 136–153.
Yang, S-Y., F-H. Sun, and J-Z. Ma, 2008: Evolvement of precipitation extremes in Northeast China on the background of climate warming (in Chinese). Sci. Geogr. Sin., 28 , 224–228.
Zhang, Y., and L. Zhang, 2005: Precipitation and temperature probability characteristics in climatic and ecological transition zone of Northeast China in recent 50 years (in Chinese). Sci. Geogr. Sin., 25 , 561–566.
Zheng, D., and D. Dong, 1986: Realization of narrow band filtering of the polar motion data with multi-stage filter. Acta Agron. Sin., 27 , 368–376.
Zheng, D., B. F. Chao, Y. Zhou, and N. Yu, 2000: Improvement of edge effect of the wavelet time–frequency spectrum: Application to the length-of-day series. J. Geod., 74 , 249–254.
Correlation of surface temperature between Changchun station and other stations over NEC (1961–2002) for (a) spring, (b) summer, (c) autumn, and (d) winter.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
As in Fig. 1, but for precipitation correlation.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
Climatological means of monthly (a) temperature (°C) and (b) precipitation (mm day−1) over Changchun station for 1909–42 and 1949–2006. The differences in temperature and precipitation between the two periods are also shown.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
Distributions of probability density function for temperature and precipitation over Changchun, for both January and July.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
(a) Monthly-mean surface temperature (°C) over Changchun. (b) Estimations of the time–frequency spectra of wavelet transform for the Changchun temperature. The red and blue colors represent the largest positive and negative amplitudes of wavelet spectra, respectively. The y coordinate represents the periodic time scales of the time–frequency spectra.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
As in Fig. 5, but for monthly-mean precipitation (mm day−1) over Changchun.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
Estimated (a),(b) cross correlation and (c),(d) squared coherence of Changchun temperature with zonal and meridional winds. In (a),(b), negative (positive) values shown in the x coordinate represent the correlations in which winds lead (lag) temperature. The dashed lines show the threshold values of significant test (assessed two-sided) at the 99% confidence level.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
As in Fig. 7, but for the correlation and coherence of precipitation with zonal and meridional winds.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
(a) Regressions of 850-mb winds (m s−1; vectors) and SST (°C; shadings) against the annual signals of Changchun temperature for the period of 1949–2006. (b) As in (a), but for the semiannual signals of Changchun temperature.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
As in Fig. 9, but for the regressions against Changchun precipitation.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
(a) Index of monthly-mean NPO. (b) Estimations of the time–frequency spectra of wavelet transform for NPO.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
Estimated (a),(b) cross correlation and (c),(d) squared coherence of Changchun temperature and precipitation with NPO. In (a),(b), negative (positive) values shown in the x coordinate represent the correlations in which NPO leads (lags) temperature or precipitation. The dashed lines show the threshold values of significant test (assessed two-sided) at the 99% confidence level. The mean annual cycles have been removed from the original data.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
(a) Estimations of coherence spectra between Changchun temperature and NPO shown in time–frequency domain. (b) Estimations of coherence spectra between Changchun precipitation and NPO. The threshold values of the significant test at the 95% and 99% confidence levels are given by the dashed–dotted lines in the color bar on the right-hand side. Note that the mean annual cycles have been removed from the original data and that a 5-yr truncation occurs at each end of the figure.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
(a) Mean patterns of annual SST (°C; shadings) and 850-mb winds (m s−1; vectors) for 1968–79. (b) Differences in annual SST and 850-mb winds (1981–92 minus 1968–79).
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
Estimated (a),(b) cross correlation and (c),(d) squared coherence of temperature and precipitation over NEC (79 stations) with NPO. Values of negative (positive) lags shown in the x coordinate represent the correlations in which temperature or precipitation leads (lags) NPO. The dashed lines show the threshold values of significant test (assessed two-sided) at the 99% confidence level. The mean annual cycles have been removed from the original data.
Citation: Journal of Climate 23, 18; 10.1175/2010JCLI3554.1
Parameter estimations of the most apparent signals calculated from the Changchun temperature for 1909–42 and 1949–2006. The estimated phases are relative to the epoch of January 1909 for data span 1909–42 and January 1949 for 1949–2006.
Parameter estimations of the most apparent signals calculated from NPO for 1948–2006. The estimated phase is relative to the epoch of January 1948.
