Impact of Autumn SST in the Japan Sea on Winter Rainfall and Air Temperature in Northeast China

SHI Xiaomeng, SUN Jilin, WU Dexing YI Li and WEI Dongni

1)College of Physical and Environmental Oceanography,Ocean University of China,Qingdao266100,P. R. China

2)Qingdao Meteorological Bureau,Qingdao266003,P. R. China

3)Dalian Marine Environmental Monitoring Division,State Oceanic Administration,Dalian116015,P. R. China

Impact of Autumn SST in the Japan Sea on Winter Rainfall and Air Temperature in Northeast China

SHI Xiaomeng1),2), SUN Jilin1),*, WU Dexing1), YI Li1), and WEI Dongni3)

1)College of Physical and Environmental Oceanography,Ocean University of China,Qingdao266100,P. R. China

2)Qingdao Meteorological Bureau,Qingdao266003,P. R. China

3)Dalian Marine Environmental Monitoring Division,State Oceanic Administration,Dalian116015,P. R. China

We studied the impact of sea surface temperature anomaly (SSTA) in the Japan Sea and the sea area east of Japan on the winter rainfall and air temperature in Northeast (NE) China using the singular value decomposition (SVD) and empirical orthogonal function (EOF). The monthly-mean rainfall data observed at 160 stations in China, monthly-mean sea surface temperature (SST) of the Hadley Center for Climate Prediction and Research and monthly-mean air temperature from the NCEP reanalysis during 1960-2011 were used. Correlation analysis indicates that the SSTAs in the Japan Sea in September may last for three or four months and are an important index for forecasting the winter rainfall and air temperature in NE China. Positive SSTAs in the central Japan Sea and in the sea area east of Tokyo correspond to positive rainfall anomaly and negative air temperature anomaly in NE China. With the rise of SST in the Japan Sea, a weak cyclone appears over the Japan Sea. The northeasterly wind transports water vapor from the Okhotsk to NE China, resulting in more rainfall and lower air temperature. Negative SSTA years are accompanied by warmer air temperature and less snow in NE China. The 1000 hPa geopotential height anomaly and wind anomaly fields are simulated by IAP-9L model, which supports the analysis results.

Japan Sea; SSTA; Northeast China; rainfall; air temperature

1 Introduction

Northeast (NE) China is located on the east coast of Eurasian continent and is a part of the East Asian monsoon system. The region has large climate variability and frequent natural hazards (Zhou, 1991). Snow, rainfall and severe cold weathers often adversely affect the region’s agriculture, industry, water conservation, and traffic, causing huge amount of economic losses. Snow disaster is often followed by a flood season. Studies on the climate change in NE China will help to improve climate forecast and reduce losses caused by weather-induced disasters.

The climate change in NE China has been extensively studied (Wanget al., 2013; Zhenget al., 2013). Trend analyses on temperature and rainfall in the northern part of NE China revealed that summer air temperature was increased during 1951-1999 (Zouet al., 2000). The variability of summer air temperature is different in the northern and southern parts of NE China on the interannual and decadal time scales (Sun and Wang, 2006). Zhanget al. (1985) found that larger area of the polar vortex corresponds to colder summer temperature in NEChina. The relationship between NE Asian (including Korea, Japan, and NE China) summer rainfall and monsoon index shows that the abnormally warm sea surface temperature (SST) in the tropical eastern Pacific can force a strong western North Pacific anticyclone in winter before a strong summer monsoon year (Leeet al., 2005). Liet al. (2010) and Sun and An (2003) discovered that the summer precipitation in NE China has a very close relation with the Pacific SST anomaly (SSTA). Time-frequency characteristics analysis shows that the regional climate change over NE China is related to thermal contrast between the Asian continent and mid/high-latitude North Pacific (Liuet al., 2010).

As one of the marginal seas in Asia, the Japan Sea (including waters around Hokkaido Island and farther southeast parts of the ocean (30°N-50°N, 125°E-150°E)) influences the climate of NE China in autumn and winter. Scholars have found that the Japan Sea SST plays an important role in changing the atmospheric circulation (Uedaet al., 2011; Iizukaet al., 2013). The increasing SST in the Japan Sea in autumn and winter help to develop the polar low over the Japan Sea (Fu, 2001), which would influence the climate variability in NE China. However, previous studies rarely focused on the relationship between winter climate variability in NE China and the Japan Sea SST.

It is usually accepted that the SST in the mid and higher latitudes is forced by the atmospheric circulation. However, some studies have also found that regional SSTs can affect regional climate in the mid-higher latitudes (Xieet al., 2003; Shi and Sun, 2009; Sunet al., 2012). Shiet al. (2013) showed that higher zonal temperature gradient in the Arctic Ocean coincides with negative NE China precipitation anomalies. Given that regional SST can affect regional climate, it is important to find out how the SSTA in the Japan Sea influences the winter climate in NE China.

In this paper, the data and methods are described in Section 2. The relationship between the SSTA in the Japan Sea (including the sea area east of Tokyo) and the winter weather in NE China is examined in Section 3. The impact and physical mechanism as well as model results on the regional ocean forcing are explored in Section 4. Discussion and conclusions are provided in Section 5.

2 Data and Methods

The SST field was derived from the monthly-mean data of the Hadley Centre for Climate Prediction and Research for the period of 1960-2011. Monthly-mean precipitation anomalies at 160 stations in China were provided by the National Meteorological Information Center. Monthly-mean air temperature, geopotential height fields and water vapor amount were obtained from the NCEP/ NCAR reanalysis data for the same period.

Statistical methods, such as singular value decomposition (SVD) and empirical orthogonal function (EOF), were used. In order to highlight large-scale patterns of covariability, we used maximum covariance analysis based on SVD of the covariance matrix between rainfall and SSTAs. SVD is widely used in meteorological research and can help to build the best collaborative change coupling model.

First, we list two standardized field SSTS(x, t) and RainfallZ(y,t). LetPkandQkbe the orthonormal vectors ofSandZ. Takingaktandbktas coefficients of the time series, the anomalies can be decomposed into

Homogeneous and heterogeneous correlation coefficients (r(St,bkt) andr(Zt,akt)) can be acquired by SVD (Wu and Wu, 2005). The homogeneous map shows that the temporal change of one mode of the field is self- affected, whose key area is the most important area of its own change (Fenget al., 2006). Brethertonet al. (1992) found that SVD is superior to the combined principal component analysis. SVD could clearly isolate the two most important extra-tropical modes of variability (Wallaceet al., 1992). SVD hss also been widely used in other studies (e.g., Juet al., 1999; Shabbar and Skinner, 2004; Danforth and Kalnay, 2008; Luet al., 2009).

3 Relationship Between SSTA in the Japan Sea and Winter Weather in NE China

3.1 SST Signal in the Japan Sea

Fig.1 Correlation coefficient between Japan Sea SSTA in September and Japan Sea SSTA in (a) October, (b) November, (c) December, and (d) January in the following year. The shaded area indicates a significant correlation (at 95% significant level byt-test).

The Japan Sea SSTA in September that persists for 3-4months is an essential prerequisite for the use of the SSTA to predict winter air temperature and rainfall in NE China. In this study, the Japan Sea SST includes SST around Hokkaido Island and farther southeast parts of the ocean, and, of course, the SST in the Japan Sea proper.

Fig.1 shows the correlation coefficients (R) between Japan Sea SSTA in September and Japan Sea SSTA from October to January (in the following year). In the following discussion,R＞ 0.3 indicates that the correlation coefficient is above the 95% significant test (P＜ 0.05), whileR＞ 0.38 shows that the correlation coefficient is above the 99% significant test (P＜ 0.01). The shaded area indicates significant correlation (R＞ 0.3,P＜ 0.05). In Fig.1a, most areas exhibit typical positive correlation (R＞ 0.38,P＜0.01), whileRreaches 0.8 in the sea east of Tokyo (34°N-39°N, 140°E-145°E). The Japan Sea SSTA in September will continue through October. The September/November correlation is positive (R＞ 0.3,P＜ 0.05) and the range is similar to that in Fig.1a, though the maximumRis reduced to 0.7 (Fig.1b). In addition, the positive correlation (R＞ 0.3,P＜ 0.05) was reduced in the seas south of the Japan Sea and to the east of Tokyo (Fig.1c). The September SSTA in the seas south of the Japan Sea and to the east of Tokyo can last three months. In Fig.1d, the key area (R＞ 0.3,P＜ 0.05) covers the southwest part of the Japan Sea and the sea area east of Tokyo. Fig.1 illustrates that the SST anomalies in the Japan Sea may last for three or four months.

3.2 Relationship Between Japan Sea September SSTA and NE China Winter Rainfall

Table 1 shows the first two SVD modes between the September SSTA in the Japan Sea and the sea area off East Japan (30.5°N-50.5°N, 125.5°E-150.5°E) and the winter rainfall anomaly in NE China. Both modes are above the 99% significance test and 95% significance test. The first mode, which is the significant mode (with higher CSCFK), will be analyzed next.

Table 1 The hetero-correlation coefficients of the first two SVD modes between the September SSTA in the Japan Seaand the winter rainfall anomaly in NE China

Fig.2 shows the results of the first SVD mode. Figs.2a, 2c, 2e show the homogeneous correlation coefficient of the SST field in September. Fig.2b, 2d, 2f show the heterogeneous correlation coefficient of the rainfall field in December, January, and February, respectively.R= 0.3 (0.38) indicates the 95% (99%) significance level. The key area in Fig.2a includes the central Japan Sea and the sea area east of Tokyo, while the key area in Fig.2b includes Heilongjiang Province, Jilin Province and eastern/central Inner Mongolia. When the SST in the central Japan Sea and the sea area east of Tokyo rises in September, the rainfall in Heilongjiang Province, Jilin Province and eastern/central Inner Mongolia will increase in December; and vice verse. The key areas in Figs.2c and 2e are similar to those in Fig.2a, while the key area in Fig.2d includes Heilongjiang Province. The key area in Fig.2f includes eastern Liaoning Province and North Korea. Positive SSTA in September in the central Japan Sea and the sea area east of Tokyo corresponds to heavy rainfall in Heilongjiang Province in January; and vice versa.

3.3 Relationship Between the Japan Sea SSTA in September and the NE China Air Temperature in Winter

From the perspective of actual climate prediction, the positive SSTA in Japan Sea and the sea area east of the Japan Sea are considered. Firstly, the September SSTA in the Japan Sea over the 51 years from 1960 to 2010 was analyzed using EOF. Based on the pattern of the first EOF mode (Fig.3a), the key area is similar to those in Figs.1 and 2. The variance contribution rate in the first mode is 55.37%. The decadal climate variability has occurred since the mid 1970s and the satellite SST data have become widely available since 1979. By designating the years with time series values greater than 100 as positive SST anomaly years, there are 13 positive years in the period 1979-2010 in Fig.3b. They are 1989, 1990, 1994, 1998, 1999, 2000, 2001, 2004, 2005, 2006, 2007, 2008, and 2010.

The December-February air temperature anomalies in the 13 years of 1979-2010 are composed. The air temperature anomaly over NE China is negative when the September SSTA in the Japan Sea is positive (Figs.4a-b). As the September SST rises in the Japan Sea, the air temperature over NE China in December and January decrease; but in February, the areas with decreasing air temperature are not in NE China. In Figs.2 and 4, the weather in NE China in December and January is significantly affected by the SSTA in the Japan Sea in September. The weather in NE China in February is affected by other factors and needs further investigation.

Fig.2 The homo/heterogeneous correlation coefficient fields of the first SVD mode between the Japan Sea SSTA in September (homogeneous fields, a, c, and e) and the NE China rainfall anomaly in December, January, and February (heterogeneous fields, b, d, and f). The shaded area indicates a significant correlation (of 95% significantt-test).

Fig.3 Pattern (a) and time series (b) of the first EOF mode of SSTA in the Japan Sea in September.

Fig.4 Composite analysis of air temperature anomaly in NE China from December to February (a-c) in significantly positive September SSTA years after 1979. The shaded area indicates negative air temperature anomaly.

4 Mechanism

4.1 Data Analysis

Figs.2-4 show that the positive SSTA in September, especially in the area between central Japan Sea and the sea area east of Tokyo, corresponds to the positive rainfall anomaly and negative air temperature anomaly in NE China in December and January. The composite analysis of wind anomaly and water vapor transport anomaly in December in the 13 years are shown in Fig.5. In Fig.5a, a weak cyclone (solid line/circle) appears over the Japan Sea and NE China. Meanwhile, abnormal northeasterly wind (Fig.5a, solid line/square frame) blows from the Okhotsk and transports water vapor to NE China (Fig.5b). In the negative anomaly years, no abnormal cyclonic circulation appears, and less moisture is transported to NE China (we omitted the map for lack of space), which easily results in warmer temperature and less snow in winter.

Fig.5 Composite analysis of (a) 1000 hPa wind anomaly and (b) 1000 hPa water vapor transport anomaly in December during prominently positive September SSTA years after 1979. The circle and square in (a) indicate the abnormal cyclone and northeasterly wind, respectively.

Snowstorm in NE China can easily cause huge agricultural and other losses. Given the need of seasonal weather forecast, the mechanism in positive SSTA years is worth exploring. This correlation can be explained by a large-scale air-sea interaction. In early winter, the land temperature declines quickly due to heat loss of earth. When the SST is abnormally high in the Japan Sea, the sea-land air temperature difference will be larger. The500hPa potential height over the region of warm sea water has positive anomaly while negative anomaly occurs over the NE China (cold land). Then the potential height contours bend to the high-latitude, forming the thermal forced trough. In the Petterssen cyclone/anticyclone development equation (Wu, 1999)

‘0’ indicates 1000 hPa while ‘5’ shows 500 hPa. The first right term is vorticity advection. Positive vorticity advection (in front of the trough) promotes the development of surface cyclone. With the collective effects of positive vorticity advection in front of the trough, surface water vapor and diabatic heating, water vapor can be easily transported from the Okhotsk to NE China, resulting in more rainfall and lower air temperature. In addition, the east part of the surface cyclone brings abnormal southerly, which can reduce the southward cold water transport from the Oyashio, thereby keeping the Japan Sea SSTA positive.

The high-altitude circulation pattern is connected with the location of the anomalous thermal wave in the mid/ high-latitude ocean of the Northern Hemisphere. In prominently positive September SSTA years, the 500 hPa geopotential height anomaly in December has two patterns： one is higher in South China and lower in North China, and the other is the opposite.

Table 2 The North Atlantic Oscillation (NAO) index and the location of cold center at 500hPa in prominently positive September SSTA years after 1979

The North Atlantic Oscillation (NAO) index and the location of cold center at 500hPa in December of the 13 years are shown in Table 2. In general, Iceland Low is strengthened when the cold center is located in North America and the NAO is positive at the same time. The 500 hPa geopotential height increases over the Bering Sea and the Bering Strait, with a blocking high in the high altitude of the Pacific. A cold eddy appears west of the blocking high and causes a 500 hPa geopotential height anomaly in December, leading to the first pattern. The negative NAO phase corresponds to lower Iceland Low, and the cold center appears in Novaya Zemlya. The planetary-scale wave over NE China is a high ridge, leading to the second pattern. Therefore, the high-altitude circulation pattern needs to be explored in future.

4.2 Model Results

Fig.6 Maps of simulated 1000 hPa geopotential height anomaly (hPa, solid black line) and 1000 hPa wind anomaly (m s-1, gray arrow) in (a) December, (b) January, (c) February. The positive SSTA remains from September to December in the Japan Sea.

The IAP-2L AGCM is the two-level atmospheric general circulation model of the Institute of Atmospheric Physics. The IAP-9L AGCM is a coupled atmosphereocean general circulation model with a horizontal resolution of 5°× 4° (Zenget al., 1987; Zhang, 1990). The model has nine unevenly spaced levels in the vertical direction with a top at 10 hPa. The moist adjustment process is as-sumed to occur in the free atmosphere. It is assumed that the PBL top has the same water vapor content as the model lowest layer. The module consists of a two-layer soil model, a surface energy balance model, and a primitive plant canopy model coupled with a PBL model. The land contours and terrain height data were provided by the Navy Fleet Numerical Oceanography Center at Monterey, US. The SST data were obtained from Alexander and Mobley (1976). Detailed description of the IAP-9L model can be found in Bi (1993). Through comparison between model results and observations, Bi (1993) showed that the model is able to simulate a realistic climate mean state. Xueet al. (2001) had succeeded in modeling the seasonal monsoon variation in the middle and lower troposphere. Wang (2002) used the IAP-9L in simulating the paleo-climate.

The 1000 hPa geopotential height anomaly and wind anomaly fields from September to February were simulated by the IAP-9L. The positive SSTA in the Japan Sea (30°N-50°N, 125°E-150°E) is offset from September to December. Taking into account the low resolution of this model, the positive SSTA are given from +3.6℃-+4.5℃in this model. Ten simulations are taken in every other 0.1℃. Fig.6 lists the numerical results of geopotential height anomaly and wind anomaly in December, January and February. In Fig.6a, negative geopotential height anomaly appears over north of the Japan Sea and abnormal wind blows from the Okhotsk and transports water vapor to NE China. In Fig.6b, the abnormal cyclone is moving northward and diminishing its impact on NE China. The positive geopotential height anomaly replaces the negative anomaly in February (Fig.6c) and NE China’s weather is unaffected by the abnormal cyclone.

Although the weak cyclone anomaly moves 3-4 degrees northeastward compared with the statistical results, the simulated anomalies in Figs.6a-c match the statistical results in Sections 3.2 and 3.3 fairly well, considering the low resolution of IAP-9L.

5 Summary and Discussion

1) Correlation analysis shows that the September SSTA in the Japan Sea can last for at least three months and even continue through the following January in the key area.

2) The September SSTA in the Japan Sea and the sea area east of Japan is an important index for forecasting the winter rainfall and air temperature in NE China. Positive SSTA corresponds to positive rainfall anomaly and negative air temperature anomaly in December and January in NE China. In the respect, the weather in NE China in February needs further investigation.

3) The relevant mechanism is as follows. The positive SSTA enhances the land-sea thermal difference, which can influence the atmospheric circulation. With the combined effect of the thermal difference and diabatic heating, a weak cyclone appears over the Japan Sea. The east part of the surface cyclone brings in abnormal southerly, which weakened the southward cold water transport from the Oyashio, thereby keeping the Japan Sea SSTA positive. The northeasterly wind transports water vapor from the Okhotsk to NE China, resulting in more rainfall and lower air temperature. The high-altitude circulation pattern is affected by the anomalous circulation of the Northern Hemisphere and ocean thermal forcing. The positive NAO phase corresponds to higher geopotential height in South China and lower geopotential height in North China, while the negative NAO phase functions in the opposite way.

Acknowledgements

The authors thank Prof. Qinyu Liu for her suggestions about the persistence of the Japan Sea SSTA. This work was supported by Innovation and Research Foundation of Ocean University of China (No. 201261009), the National Natural Science Foundation of China (Nos. 40930844 and 10735030) and the National Basic Research Program of China (the 973 Program) under grant No. 2005CB422 301.

Alexander, R. C., and Mobley, R. L., 1976. Monthly average sea-surface temperature and ice-pack limits on a 1° global grid.Monthly Weather Review, 104： 143-148.

Bi, X. Q., 1993. IAP 9-level atmospheric general circulation model and climate simulation. PhD thesis. Institute of Atmospheric Physics, Chinese Academy of Sciendes, Beijing (in Chinese with English abstract).

Bretherton, C. S., Smith, C., and Wallace, J. M., 1992. An intercomparison of methods for finding coupled patterns in climate data.Journal of Climate, 5： 541-560.

Danforth, C. M., and Kalnay, E., 2008. Using singular value decomposition to parameterize state-dependent model errors.Journal of the Atmosphere Sciences, 65： 1467-1478, DOI：10.1175/2007JAS2419.1.

Feng, X., Wang, X., and Wang, Y., 2006. Anomalies of the Northeast China floods season precipitation and SVD analysis with SSTA in world Oceans.Journal of Tropical Meteorology, 22 (4)： 367-373 (in Chinese with English abstract).

Fu, G., 2001.Polar Lows: Intense Cyclones in Winter. China Meteorological Press, Beijing, 164-172.

Iizuka, S., Shiota, M., Kawamura, R., and Hatsushika, H., 2013. Influence of the monsoon variability and sea surface temperature front on the explosive cyclone activity in the vicinity of Japan during Northern Winter.SOLA, 9 (0)： 1-4.

Ju, J. H., Deng, S., Chen, X. F., and Yan, H. S., 1999. Field correlation analysis between the monthly-mean 500 hPa height anomaly from January to May and rainfall of china in summer.Journal of Tropical Meteorology, 15： 154-161 (in Chinese with English abstract).

Lee, E. J., Jhun, J. G., and Park, C. K., 2005. Remote connection of the Northeast Asian summer rainfall variation revealed by a newly defined monsoon index.Journal of Climate, 18：4381-4393.

Li, F., Li, J., and Guan, Z. Y., 2010. Inter-decadal variations of summer temperature in Northeast China and relationships with Pacific SSTA.Journal of Meteorology and Environment, 26 (3)： 19-26 (in Chinese with English abstract).

Liu, S., Yang, S., Lian, Y., Zhang, D. W., Wen, M., Tu, G., Shen,B. Z., Gao, Z. T., and Wang, D. H., 2010. Time-frequency characteristics of regional climate over Northeast China and their relationships with atmospheric circulation patterns.Journal of Climate, 23： 4956-4972.

Lu, C. H., Guan, Z. Y., Wang, P. X., and Duan, M. K., 2009. Detecting the relationship between summer rainfall anomalies in Eastern China and the SSTA in the global domain with a new significance test method.Journal of Ocean University of China, 8： 15-22.

Shabbar, A., and Skinner, W., 2004. Summer drought patterns in Canada and the relationship to global sea surface temperatures.Journal of Climate, 17： 2866-2880.

Shi, D. D., and Sun, J. L., 2009. Study on seasonal variation of heat content in marginal seas in the east of China.Periodical of Ocean University of China, 39： 274-280 (in Chinese with English abstract).

Shi, X. M., Sun, Y. W., and Sun, J. L., 2013. The impact of zonal temperature gradient in the Arctic Ocean on the summer precipitation over Northeast China.Periodical of Ocean University of China, 44 (2)： 11-16 (in Chinese with English abstract).

Sun, L., and An, G., 2003. The effect of north Pacific sea surface temperature anomaly on the summer precipitation in Northeast China.Acta Meteorological Sinica, 61 (3)： 346-353 (in Chinese with English abstract).

Sun, J. L., Cong, M., Wu, D. X., and Gao, S. H., 2012. The effect of meridional thermal difference in eastern marginal seas of China to climate change in Nanjing during summer.Periodical of Ocean University of China, 42 (5)： 001-006 (in Chinese with English abstract).

Sun, J. Q., and Wang, H. J., 2006. Regional difference of summer air temperature anomalies in Northeast China and its relationship to atmospheric general circulation and sea surface temperature.Chinese Journal of Geophysics,49 (3)： 588-598.

Ueda, A., Yamamoto, M., and Hirose, N., 2011. Meteorological influences of SST anomaly over the East Asian marginal sea on subpolar and polar regions： A case of an extratropical cyclone on 5-8 November 2006.Polar Science, 5 (1)： 1-10.

Wallace, J. M., Smith, C., and Bretherton, C. S., 1992. Singular value decomposition of wintertime sea surface temperature and 500-mb height anomalies.Journal of Climate, 5： 561-576.

Wang, H., Liu, Q. Y., and Zheng, J., 2013. Formation mechanism for the anomalous anticyclonic circulation over Northeast Asia and the Japan Sea in boreal winter 1997/98 and the spring of 1998.Journal of Ocean University of China, 12 (2)：312-317.

Wang, H. J., 2002. The Mid-Holocene climate simulated by a grid-point AGCM coupled with a biome model.Advances in Atmospheric Sciences, 19 (2)： 205-218.

Wu, H. B., and Wu, L., 2005.Climate Variability Diagnosis and Prediction Methods. China Meteorological Press, Beijing, 103-163 (in Chinese).

Wu, R. S., 1999.Principles of Modern Synoptic Meteorology. Higher Education Press, Beijing, 136-140 (in Chinese).

Xie, S. P., Xie, Q., Wang, D., and Liu, W. T., 2003. Summer upwelling in South China Sea and its role in regional climate variations.Journal of Geophysical Research-Oceans, 108 (C8)： 17.1-17.13, DOI： 10.1029/2003JC001867.

Xue, F., Bi, X. Q., and Lin, Y. H., 2001. Modeling the global monsoon system by IAP 9L AGCM.Advances in Atmospheric Sciences, 18 (3)： 404-412.

Zeng, Q. C., Yuan, C. G., Zhang, X. Z., Liang, X. Z., and Bao, N., 1987. A global gridpoint general circulation model.Collection of Paper Presented at the WMO/IUGG NWP Symposium, Tokyo, 421-430.

Zhang, S. Q., Yu, T. J., Li, F. Y., Wang, X. M., Wang, X. F., and Wu, W. M., 1985. The seasonal variations of area and intensity of polar vortex in northern hemisphere and relationship with temperature in northeast China.Scientic Atmospherica Sinica, 9 (2)： 178-185 (in Chinese with English abstract).

Zhang, X. H., 1990. Dynamial framework of IAP fine-level atmospheric general circulation model.Advances in Atmospheric Sciences, 7 (1)： 67-77.

Zheng, H. F., Mclaughlin, N. B., He, X. Y., Yu, X. Y., Ren, Z. B., and Zhang, D., 2013. Temporal and geographical variation in the onset of climatological spring in Northeast China.Theoretical and Applied Climatology, 114 (3-4)： 605-613, DOI： 10.1007/s00704-013-0869-1.

Zhou, L., 1991.Climate of Northeast China. China Meteorological Press, Beijing, 125pp (in Chinese).

Zou, L. Y., Ma, J. X., and Zhou, J. L., 2000. Preliminary study on trends of temperature and precipitation in the north of the northeast China.Journal of Nanjing Institute of Meteorology, 23 (4)： 560-567 (in Chinese with English abstract).

(Edited by Xie Jun)

(Received September 10, 2013; revised October 15, 2013; accepted May 30, 2015)

? Ocean University of China, Science Press and Spring-Verlag Berlin Heidelberg 2015

* Corresponding author. Tel： 0086-532-66782127 E-mail： rainbetimes@163.com