Changing characteristics and spatial differentiation of spring precipitation in Southwest China during 1961–2012<xref ref-type="fn" rid="cpb141474fn1">*</xref>

Cite this Article

Liu Hong-Lan, Zhang Qiang, Zhang Jun-Guo, Hu Wen-Chao, Guo Jun-Qin, Wang Sheng. Changing characteristics and spatial differentiation of spring precipitation in Southwest China during 1961–2012* . Chinese Physics B, 2014, 24(2): 029201 Copy to clipboard

Permissions

Changing characteristics and spatial differentiation of spring precipitation in Southwest China during 1961–2012*

Liu Hong-Lan^a)†, Zhang Qiang^b), Zhang Jun-Guo^c), Hu Wen-Chao^d), Guo Jun-Qin^e), Wang Sheng^f)

a)Zhangye Meteorological Bureau, Zhangye 734000, China

b)Institute of Arid Meteorology, China Meteorological Administration, Key Laboratory of Arid Climatic Change and Disaster Reduction of Gansu Province, Key Laboratory of Arid Climate Change and Disaster Reduction of China Meteorological Administration, Lanzhou 730020, China

c)Zhangye Middle School, Zhangye 734000, China

d)Meteorological Service Center of Gansu Province, Lanzhou 730020, China

e)Northwest Regional Climate Center, Lanzhou 730020, China

f)Jinta County Meteorological Bureau, Jinta 735300, China

†Corresponding author. E-mail: gszylhl@126.com

*Project supported by the National Basic Research Program of China (Grant No. 2013CB430200 (2013CB430206)) and the Sixth Program Ten Talented People of the Meteorological Bureau of Gansu Province, China.

Abstract

In this study, we analyze spring precipitation from 92 meteorological stations spanning between 1961 and 2012 to understand temporal–spatial variability and change of spring precipitation over Southwest China. Various analysis methods are used for different purposes, including empirical orthogonal function (EOF) analysis and rotated EOF (REOF) for analyzing spatial structure change of precipitation anomaly, and the Mann–Kendall testing method to determine whether there were abrupt changes during the analyzed time span. We find that the first spatial mode of the precipitation has a domain uniform structure; the second is dominated by a spatial dipole; and the third contains five variability centers. The 2000s is the decade with the largest amount of precipitation while the 1990s is the decade with the smallest amount of precipitation. The year-to-year difference of that region is large: the amount of the largest precipitation year doubles that of the smallest precipitation year. We also find that spring precipitation in Southwest China experienced a few abrupt changes: a sudden increase at 1966, a sudden decrease at 1979, and a sudden increase at 1995. We speculate that the spring precipitation will increase gradually in the next two decades.

Keyword: 92.05.Df; 92.40.Cy; 92.40.Ea; spring precipitation; temporal–spatial structure; 500-hPa level; Southwest region

Show Figures

1. Introduction

China is one of the most vulnerable countries in the world to natural disasters: the annual loss mounts to 3%– 6% of the gross domestic product caused by weather/climate disasters.^[1] The frequent occurrences of drought have been threatening food production, social security, thereby creating tremendous pressure and challenges to the socio-economic conditions and environment-ecology problems. Southwest China (97° 21′ – 110° 12′ E, 21° 08′ – 34° 19′ N) is located in Eastern Asia, and includes the Sichuan Province, Chongqing City, Yunnan Province, and Guizhou Province, spreading over 1.12 million square kilometers. The region contains the western Sichuan Plateau, Sichuan basin, Yungui Plateau, Henduan Montain, etc, and is orographically complicated and features diverse climates. The yearly mean precipitation in that region ranges from 600 mm to 1200 mm, decreasing from its southeastern part to its northwestern part, largely depending on the orographic distribution that affects the penetration of the warm and wet southeasterly.

Partly associated with global climate change, China has experienced numerous regional weather/climate extreme events since the 1980s, especially frequent severe drought in its southern regions.^[2] For example, the 2003 summer– autumn drought occurred in the Southeast, South, and part of Southwest China; ^[3] the 2005/2006 autumn– winter– spring drought in the southern part of South China, and 2005’ s was the most severe spring drought in the Yunnan province for the last half century; 2006’ s record setting summer drought in Sichuan Province and Chongqing City; ^{[4, 5]} the 2007/2008 winter– spring– summer drought and 2009’ s large scale autumn– winter– spring drought in most of Southwest China. These droughts brought extreme difficulties to agriculture, industry, and people’ s lives.^{[6– 8]} He et al.^[9] reported characteristics of extreme drought climate change in Southwest China over nearly 50 years; Zheng et al.^[10] analyzed the spring abnormal circulation of extreme drought years in Yunnan; Li et al.^[11] reported the vapor transport of drought and flood in the summer in Southwest China; Xu et al.^[12] studied the asymmetric relation between winter North Atlantic Oscillation and the precipitation in Southwest China; Li et al.^[13] reported spatial and temporal evolution characteristics of summer precipitation in the east of Southwest China; Zhang et al.^[14] analyzed the main mode of precipitation in dry or wet season in Southwest China. Understanding the climate variability of Southwest China and developing and utilizing climatic resources, especially the precipitation variability in the spring season (March to May) when summer harvesting plants (rice, tobacco, corn) experience their critical growing stages, are likely to bring tremendous benefits. These reasons motivate us to study the climate variability and change in Southwest China, which could potentially project the vital skill of speculating the future climate variability and changes of that region.^{[15, 16]}

2. Data and method

Southwest China mentioned in this study refers to the territory of China, as enclosed by a rectangle from 97° 21′ E to 110° 12′ E, from 21° 08′ N to 34° 19′ N, as shown in Fig. 1. This part of China includes the Sichuan Province, Chongqing City, Yunnan Province, and Guizhou Province. The data analyzed include spring (from March to May) precipitation from 92 meteorological stations (with their spatial locations marked in Fig. 1) in the period 1961– 2012, 500-hPa level geopotential height from NCEP/NCAR reanalysis with a resolution of 2.5° × 2.5° .

	Figure Option View Download New Window
	Fig. 1. Locations of meteorological stations (dots) providing precipitation data.

The empirical orthogonal function (EOF) analysis and rotated EOF (REOF) are used to determine the basic characteristics of the spring precipitation in Southwest China as well as its spatiotemporal structures.^[17] The Mann– Kendall testing method is used to determine the temporal locations of abrupt changes of spring precipitation, ^[17] with a priori set a = 0.05 significance level. The trend is determined using the fifth-order running trend, and its statistical significance is also tested using a non-parametric Mann– Kendall method. The periodicity information of the spring precipitation is obtained using spectrum analysis.^[17] A t-testing is used to test the statistical significance of abrupt changes. Finally, the mean generation function method is used to construct a short-term climate prediction model for the spring precipitation in Southwest China.^[17]

3. Precipitation characteristics in Southwest China

3.1. Basic climate characteristics

Southwest China has a wide spread of spring precipitation from 54.8 mm to 581.0 mm, with a mean value of 213.5 mm. Spring precipitations range from 6.9% to 33.6% of the annual total for different regions, and a mean value is 20.4%. Figure 2 shows the plots of the mean and the standard deviations of precipitations over the whole Southwest China region. It is evident that the southeastern part has abundant rainfall in spring while the northwestern part has much less rainfall. The dry center is located at Batang County (98° 58′ E, 28° 46′ N) of western Sichuan Province, and the spring precipitation is only 54.8 mm. The largest spring precipitation occurs in the neighboring regions of Gongshan Derung ang Nu Autonomous County (98° 40′ E, 27° 45′ N) of Yunnan Province, and its value is over 580 mm. A secondary local maximum center is located at Tongren County (109° 11′ E, 27° 43’ N) of Guizhou Province, and its value is 402.3 mm. The large ratio (> 10) between the values of maximum and minimum precipitation indicates the nonuniformity of the precipitation distribution over Southwest China. The standard deviation of the precipitation has a similar spatial structure of the precipitation itself, and its maximum is located at Gongshan Derung ang Nu Autonomous County (98° 40′ E, 27° 45′ N) of Yunnan Province (184.1 mm), and a minimum at Batang County (98° 58′ E, 28° 46′ N) of Sichuan Province (24.8 mm). In general, the southern and eastern parts of Southwest China are abundant rainfall areas; the remaining parts of Southwest China are lower rainfall areas. The eastern wet tongue (maximum precipitation region) extends northwestward, passing through the neighborhood of E’ mei Montain (103° 20′ E, 29° 31′ N), to the Xiaojin County (102° 21′ E, 31° 00′ N) of the Aba Tibetan Autonomous Region of Sichuan Province. Such a spatial structure implies that the precipitation of Southwest China may be highly affected by the water vapor transportation driven by the Southeastern surrounding the Pacific Subtropical High, and thereby strengthening (weakening) precipitation over that region in the years of strong (weak) water vapor convergence and transportation.

	Figure Option View Download New Window
	Fig. 2. Long-term mean spring precipitation (a) and its standard deviation (b) in Southwest China (in mm).

3.2. Spatial structure of precipitation anomaly

Table 1 presents the variance contribution from the first eight eigenvectors (empirical orthogonal functions). It is shown that the convergence of the sum of eigenvectors for the precipitation in Southwest China is relatively slow. The accumulated variance contribution from the first two eigenvectors is 33.5%, the first four 47.9%, and the first eight 63.2%. The first two eigenvectors are presented in Fig. 3. The first eigenvector (EOF) has positive values all over the domain, with a maximum value at Huili County ((102° 15′ E, 26° 39′ N) of Southern Sichuan. The result implies that Southwest China may be considered as one natural climatological regime for the trend of the spring precipitation over Southwest China, with its representative station being Huili County (102° 15′ E, 26° 39′ N) of Sichuan Province, which has a value of 0.82. The second eigenvector is dominated by a dipole, with zero-line tilting northwest-southeastward and crossing E’ Mei Mountain (103° 20′ E, 29° 31′ N), Xichang region (102° 16′ E, 27° 54′ N) of Sichuan Province, and Luxi County (103° 146′ E, 24° 32′ N) of Yunnan Province. To the east of the zero line, the representative station is the Anshun region (105° 54′ E, 26° 15′ N) of Yunnan Province with a value of 0.70; and to the west, the representative station is Ganzi Tibetan Autonomous prefecture (100° 00′ E, 31° 37′ N) of Sichuan Province with a value of − 0.41, implying that the trend of Southwest China spring precipitation has a component in contrast to the changes in the east and west parts of the Southwest China.

Table 1. Different EOF variance contributions of spring precipitation in Southwest China.

	Figure Option View Download New Window
	Fig. 3. The first (a) and second (b) characteristic vectors of spring precipitation in Southwest China.

After rotating the first eight vectors to maximize the explained variance and summing the first five rotated eigenvectors (with a total explained variance of 52.5%), five distinctive natural climatological regimes are obtained: western Sichuan Plateau, Sichuan Basin, Hengduan Mountain region, Yunnan Plateau, and eastern Guizhou Hilly Area.

3.3. Interannual and decadal variability of the spring precipitation in Southwest China

To quantify the variation of spring precipitation in Southwest China, we calculate the percentages of positive anomaly and negative anomaly. There appeared a positive anomaly in the data of a 25-year temporal span, and a negative anomaly was observed in the data of a 25-year temporal span. The interannual variability of spring precipitation is large, e.g., the fluctuation in 1990 reached a maximum value of 270.3 mm while the fluctuation in 1979 was as small as 149.8 mm. The maximum interannual fluctuation almost doubles that of the minimum. From Fig. 4, it can be identified that the fluctuating spring precipitation in Southwest China has decreased slowly in the last 52 years. Further analysis of mean anomalies over different decades (with respect to the mean in the period of 1981– 2010 over the whole Southwest China) is presented in Fig. 5. In the 1960s, the positive anomalies were located in eastern Sichuan Province, Guizhou Province, and most part of Chongqing City, while negative anomalies dominated other areas. The most evident increase of precipitation appeared in the Fengjie region (109° 32′ E, 31° 01′ N) of northeast Chongqing City: the positive anomaly reached 27%. In contrast, in the central area of Yunnan the negative anomaly was 38%. In the 1970s, Chongqing City, Guizhou Province, northern Sichuan Province, and eastern Yunnan Province each were dominated by a positive anomaly, other areas were each dominated by a negative anomaly. The increase of spring precipitation was most evident in eastern Yunnan Province and southwestern Guizhou Province: the positive anomaly reached 28%. In contrast, the central Yunnan Province (in the vicinity of Dali City) had the most evident decrease of precipitation: the negative anomaly was 25%. In the 1980s, eastern and western parts of Southwest China each had a negative anomaly and other areas each had a positive anomaly. The increase of spring precipitation was most evident in the Zhaotong region (103° 43′ E, 27° 21′ N)– Yibin region (104° 34′ E, 28° 42′ N) of northeastern Yunnan Province: the positive anomalies range from 15% to 20%. In contrast, the decrease of spring precipitation occurred mostly in the western Sichuan Province and central Yunnan Province: the negative anomaly reached 28%. In the 1990s, the northwest part of Southwest China, southeastern and northwestern parts of Yunnan province, and northeastern part of Chongqing City each were dominated by a positive anomaly, and other areas each were dominated by a negative anomaly. The most evident increase of positive spring precipitation occurred in the Shangri-La region (99° 42′ E, 27° 50′ N) of northwest Yunnan Province: the positive anomaly was 18%. In contrast, the precipitation decrease was most evident in the southern Yunnan Province: the negative anomaly was 23%. In the 2000s, northwestern Sichuan Province, northwestern Yunnan Province, and the border region between Yunnan Province and Guizhou Province and its extension to the Suining region (105° 33′ E, 30° 30′ N) of Sichuan Province each were dominated by a positive anomaly and other areas each were dominated by a negative anomaly. The precipitation increase is most evident in the central Yunnan Province: its positive anomaly reached 37%. In contrast, the northeastern and northwestern areas of Yunnan Province had the most evident decrease of spring precipitation: the negative anomaly was 10%.

	Figure Option View Download New Window
	Fig. 4. Variations of spring precipitation with time in Southwest China.

	Figure Option View Download New Window
	Fig. 5. Percentages of spring precipitation anomaly in Southwest China in the (a) 1960s, (b) 1970s, (c) 1980s, (d) 1990s, and (e) 2000s.

In summary, the 2000s was the decade of the largest quantity of spring precipitations in the last half century in Southwest China. The 1960s were dominated by above normal spring precipitation in the northeastern part and below normal spring precipitation in the southwestern part. In the 1970s, the eastern part had above normal precipitation, and the western part below normal precipitation. The 1980s featured below normal precipitation in both the eastern and western parts but above normal in the central parts. The 1990s featured above normal precipitation in the northwestern part and below normal elsewhere.

3.4. Multiscale analysis of spring precipitation in Southwest China

Linear fitting to the spring precipitation anomaly in Southwest China shows a slow steady decrease of spring precipitation, with a slope of − 0.01. From Fig. 4, it is identified that the spring precipitations in the periods of 1967– 1983 and 1998– 2009 were above normal; whereas those in the periods of 1961– 1966 and 1984– 1997 were below normal. The above normal spring precipitation in the first decade of the 21st century shows that the spring precipitation has increased.

Morlet wavelet analysis of spring precipitation in Southwest China has been characterized by multiscale variability.^[17] The dominant timescales of variability are 15 years and 25 years. The 15-year variability is evident through the whole temporal domain of analysis, i.e., 1961– 2012; it has a below-above-below-above-below-above-below normal pattern, the spring precipitations in the time periods of 1961– 1969, 1978– 1984, 1993– 1999, 2007– 2012 were below normal. The spring precipitation in 2012 was close to normal, implying an above normal precipitation in the coming years. Between these periods, the spring precipitation was above normal. The 25-year cycle appears to be most dominant, with a below-above-below-above-below normal pattern. For that cycle, the time periods of 1961– 1969, 1983– 1994, and after 2009 featured below normal spring precipitation. Between these periods, the spring precipitation corresponding to this cycle was below normal. The variability of timescales shorter than 15-years does not show any regular cycle, for the variability of each of the shorter timescales is very noisy.

	Figure Option View Download New Window
	Fig. 6. Wavelet analysis of spring precipitation in Southwest China.

Spectral analysis of the spatial standard deviation of spring precipitation in Southwest China shows peaks of 6-year, 8-year, and 15-year in the spectral diagram, which is presented in Fig. 7.

	Figure Option View Download New Window
	Fig. 7. Spectrum analysis of the spatial standard deviation (F) of spring precipitation in Southwest China.

4. Abrupt changes of spring precipitation in Southwest China

The Mann– Kendall test method and moving t-test method are used to detect the abrupt changes of spring precipitation in Southwest China.

To determine whether the spring precipitation in Southwest China has experienced abrupt changes, we use the Mann– Kendall testing method to analyze area averaged precipitation series.^[17] The Mann– Kendall testing method is a currently widely used non-parametric method. It does not assume any a priori distribution of sample and is not sensitive to a small number of abnormal extremes. It can be applied to both categorized data and series with intuitively simple calculations. One of the advantages of this method is its capability of determining the locations of mean or slope changes; and therefore, is a method that can detect abrupt changes. Figure 8 presents the detecting results, showing that the spring precipitation started to increase at the mid 1960s and was maintained until 1980. From 1980, the spring precipitation began to decrease. From the mid 1990s, the spring precipitation began to increase again. These changes appear to be abrupt, and 1966, 1979, and 1995 are considered to be the temporal locations with the abrupt changes. All these changes are statistically significant with a considerable level of a = 0.05 (u_0.05 = ± 1.96).

	Figure Option View Download New Window
	Fig. 8. Abrupt changes of spring precipitation in Southwest China, obtained from (a) the Mann– Kendall test and (b) the moving t-test, respectively.

The moving t-test method is used to detect the abrupt changes of spring precipitation from 1961 to 2012. Here assume that n = 52, two sub-sequence lengths n₁ = n₂ = 10, the significant level α = 0.05, and t_0.05 = ± 2.10, then the sequence of the t statistic will be calculated and mapped together with t_0.05 = ± 2.10 (Fig. 8(b)). From the graph, it can be seen that the t statistic is more than the level of significance 0.05 twice from the 1970s. One is positive (in 1978), and the other is negative (in 1995). This indicates that there have been two abrupt changes of spring precipitation in Southwest China in the past 52 years. It experienced a change from more to less in the 1970s, and a change from less to more in the 1990s.

5. Abnormal response of spring precipitation to 500-hPa level geopotential height

Spring is an interim season with large precipitation variability and few prediction skills. It contains both the signatures of winter circulation and summer circulation. The interplay of these two types of circulation makes spring circulation have dramatic changes. For this reason, analyzing the abnormal circulation in spring is critical to understand the structure and evolution of spring precipitation.^{[18– 20]}

The analysis of the relationship between precipitation and geopotential height can help to reveal how the synoptic system affects spring precipitation in Southwest China. As shown by the composite 500-hPa level geopotential maps in Fig. 9, in which figure 9(a) shows the rainy spring scenario and figure 9(b) displays the droughty spring scenario, the negative height anomaly in Eurasia extends from the Ural Mountains to the Mediterranean. Most Asian regions north of 38° N have a positive height anomaly. Lake Baikal features the center of 3 dagpm. A negative anomaly extends to the south, and a negative center extends from the east Bay of Bengal to the Indochina Peninsula. These features constitute a pattern that is west-low and east-high in high latitude Eurasia and north-high and south-low in Asia. The corresponding surface pressure field (not shown here) displays a vast region with a positive surface pressure anomaly (east to 90° E with its center located in the vicinity of Lake Baikal and Mongolian). For the droughty years, in contrast, the 500-hPa level geopotential map (Fig. 9(a) shows the rainy spring scenario and figure 9(b) displays the droughty spring scenario) features a positive anomaly in the polar region, a wide spread negative anomaly region covering mid- and high-latitude areas in the Atlantic, Europe, and Asia, with a central anomaly lower than 5 dagpm, and a positive anomaly center over Europe and Asia regions south to 40° N. The surface map shows a negative anomaly region covering most of Asia south to Lake Baikal, with its center located at northwest China.

These analyses show that there are major differences in the circulation pattern between rainy spring year and droughty spring year. For the rainy spring year, geopotential height anomaly features a pattern that is west-low and east-high in mid- and high-latitude areas of Eurasia, a north– south dipole in Asia with a positive anomaly to the north of 38° N and a negative anomaly to the south. Corresponding to these patterns, the mid-latitude area has an easterly anomaly and a weakened westerly zone, and the major circulation has a strong meridional component in eastern Asia, which is favorable for the cold surge to frequently move southward. For the droughty spring year, the geopotential height anomaly has a more zonal pattern, with positive centers located in the polar region and south to 40° N and a negative anomaly in between. This pattern of geopotential height anomaly corresponds to a westerly anomaly that enhances the westerly jets. Under such a kind of circulation, Siberia High is relative weak and the cold surge can hardly move to Southwest China. It is clear that in the rainy spring year, anomalous cyclonic circulation extends from the Indochina Peninsula to Southwest China, and Southwest China has an easterly anomaly. In the droughty year, the Bay of Bengal, Indochina Peninsula, and South China Sea feature anomalous anticyclonic circulation, a weakened southern trough, and a westerly or northerly anomaly.

	Figure Option View Download New Window
	Fig. 9. 500-hPa geopotential height anomaly distribution of spring in the rainy year (a) and droughty year (b) in Southwest China

6. Projection of future spring precipitation in Southwest China

The coupled climate model projection of future spring precipitation in western China by Zhang et al.^[21] predicts an increasing precipitation in the next eight decades. This increasing trend starts with little change in the first decade, and will have about 15% precipitation increase in the later years of the 80 year projection. By 2050, the largest precipitation area will have been in Southwest China, with an annual precipitation increase being over 200 mm.

In this study, we use a mean generating function to construct a spring precipitation model for Southwest China. Following the study by Wei et al., ^[17] for a given time series of n samples

the mean generating function is defined as

where i = 1, 2, … , l; l = 1, 2, … , m; n_l = int(n/l); m = int(n/2) is the maximum possible period.

The cyclic extrapolation of x_l(i) leads to

where t = 1, 2, … , N+ q; q is the prediction span. In this way, one obtains M periodic series of length N + q. Using these time series as predictors for a multivariable regression model, we obtain the prediction value of n+ q location, i.e.,

In Eq. (4), the coefficient is obtained using a least-square method.

We use the 1961– 2002 period data to determine the model parameters. The model is then used to make retrospective predictions of spring precipitation for an independent period 2003– 2012. As shown in Fig. 10, the model can fit the model training data with a 78.6% accuracy, and predict independent verification period with a 80.0% accuracy. The model error is large in the 1960s and the fitting accuracy has improved since the 1970s.

	Figure Option View Download New Window
	Fig. 10. Projection of future spring precipitation in Southwest China (in mm).

The future spring precipitation in Southwest China is predicted to have a slow steady increase. This result implies that the future increase of precipitation, as predicted by Zhang et al., ^[21] is season-dependent; and main contribution to the increase of future precipitation in Southwest China comes mainly from the increased precipitations in summer, autumn, and winter seasons, but not in the spring season.

7. Conclusions and discussion

(i) The spring precipitation in Southwest China is extremely nonuniform. The driest area is located at Batang of western Sichuan Province: the spring precipitation is only 54.8 mm. The largest spring precipitation occurs in the neighboring regions of Gongshan of Yunnan Province: its value is over 580 mm. Southern and eastern parts of that region are abundant in precipitation while other areas are dry. The first EOF has the same sign values all over the domain; the second EOF is dominated by a dipole, implying the presence of two natural climatological regimes; and the third EOF implies five distinctive natural climatological regimes: Western Sichuan Plateau, Sichuan Basin, Hengduan Mountain region, Yunnan Plateau, and eastern Guizhou Hilly Area.

(ii) The spring precipitation in Southwest China has a strong interannual and decadal variability. The largest interannual fluctuation occurred in 1990 with a value of 270.3 mm while the smallest fluctuation was in 1979 with a value of 149.8 mm. The maximum interannual fluctuation almost doubles that of the minimum. Cyclic changes appear to exist on different timescales, with a dominant pattern of below-above-below-above-below-above-below normal precipitation. Spectrum analysis shows that the large amplitude cycles include those of 6-year, 8-year, and 15-year periods.

(iii) From a decadal perspective, the 2000s is the decade of the largest quantity of precipitation while the 1990s is the decade of the smallest quantity of precipitation. There were three abrupt changes of spring precipitation from 1961 to 2012. 1966 was the year that had an abrupt increase; 1979 had an abrupt decrease; and the 1995 had an abrupt increase again. All these abrupt changes have a statistical significance of the a = 0.05 (u_0.05 = ± 1.96) level.

(iv) There are major differences in the circulation pattern between an abundant precipitation year and a dry year. In the abundant precipitation year, 500-hPa geopotential height anomaly features a west-low east-high pattern in mid- and high-latitude areas in Eurasia and a pattern of north-positive south-negative with respect to 38° N in Asia. In the dry year, 500-hPa geopotential height anomaly features only one north-negative south-positive pattern in both Europe and Asia, with a weakened southern trough. The spring precipitation in Southwest China will slowly increase in the next two decades.

(v) There remain other factors that affect the spring precipitation in Southwest China which need to be investigated. What are the roles they play? Which factor is the key factor? Numerical simulations with sensitivity experiments may help to obtain the answers. These are our future works.

Reference

View Option

1	Liu H L, Zhang Q, Hu W C and Guo J Q 2013 Journal of Glaciology and Geocryology 35 857 DOI:10.7522/j.issn.1000-0240.2013.0097 [Cited within:1] [CJCR: 1.301]
2	Liu H L, Li D L and Guo J Y 2004 Journal of Glaciology and Geocryology 26 55 [Cited within:1] [CJCR: 1.301]
3	Liu Y, Zhao E X, Huang W, Sun D and Ju J H 2007 Journal of Tropical Meteorology 23 35 [Cited within:1] [CJCR: 1.371]
4	Li Y H, Xu H M and Liu D 2011 Acta Meteorologica Sinica 25 176 DOI:10.1007/s13351-011-0025-8 [Cited within:1] [JCR: 0.799] [CJCR: 0.363]
5	Cen S X, Gong Y F, Qin N S, Lai X, Wang X and Huang X L 2011 Acta Meteorologica Sinica 69 1009(in Chinese) [Cited within:1] [JCR: 0.799] [CJCR: 0.363]
6	Yang H, Song J, Yan H M and Li C Y 2012 Climatic and Environmental Research 17 315(in Chinese) [Cited within:1] [CJCR: 1.166]
7	Song J, Yang H and Li C Y 2011 Chinese Journal of Atmospheric Sciences 35 1009 DOI:10.3878/j.issn.1006-9895.2011.06.02 [Cited within:1] [CJCR: 1.948]
8	Duan H X, Wang S P and Feng J Y 2010 Journal of Arid Meteorology 28 238(in Chinese) [Cited within:1]
9	He J Y, Zhang M J, Wang P, Wang S J and Wang X M 2011 Acta Geographica Sinica 66 1179(in Chinese) [Cited within:1] [CJCR: 2.937]
10	Zheng J M, Zhang W C, Wan Y X and Duan X 2013 Plateau Meteorology 32 1665 DOI:10.7522/j.issn.1000-0534.2012.00143 [Cited within:1] [CJCR: 1.688]
11	Li Y H, Xu H M, Gao Y H and Li Q 2010 Acta Meteorologica Sinica 68 932(in Chinese) [Cited within:1] [JCR: 0.799] [CJCR: 0.363]
12	Xu H L, Li J P, Feng J and Mao J Y 2012 Acta Meteorologica Sinica 70 1276(in Chinese) [Cited within:1] [JCR: 0.799] [CJCR: 0.363]
13	Li Y H, Xu H M, Bai Y Y and Li Q 2010 Plateau Meteorology 29 523(in Chinese) [Cited within:1] [CJCR: 1.688]
14	Zhang W L, Zhang J Y and Fan G Z 2014 Chinese Journal of Atmospheric Sciences 38 590(in Chinese) [Cited within:1] [CJCR: 1.948]
15	Yin H and LI Y H 2013 Journal of Arid Meteorology 31 182(in Chinese) [Cited within:1]
16	Dai X G, Wang P and Zhang K J 2012 Chin. Phys. B 21 119201 DOI:10.1088/1674-1056/21/11/119201 [Cited within:1] [JCR: 1.148] [CJCR: 1.2429]
17	Wei F Y 2007 Modern Technology of Climate Statistics and Diagnose2nd edn Beijing Meteorological Press [Cited within:7]
18	Zhang W and Wan S Q 2008 Chin. Phys. B 17 2311 DOI:10.1088/1674-1056/17/6/063 [Cited within:1] [JCR: 1.148] [CJCR: 1.2429]
19	Gao X Q and Zhang W 2005 Chin. Phys. 14 2370 DOI:10.1088/1009-1963/14/11/039 [Cited within:1] [JCR: 0.811] [CJCR: 0.4541]
20	Zhang D Q, Feng G L and Hu J G 2008 Chin. Phys. B 17 736 DOI:10.1088/1674-1056/17/2/062 [Cited within:1] [JCR: 1.148] [CJCR: 1.2429]
21	Zhang Y J, Dong W J, Yu Y G and Feng J M 2004 Climatic and Environmental Research 9 342(in Chinese) [Cited within:2] [CJCR: 1.166]

2013

0.0

1.301

Liu

H L

, Zhang

, Hu

W C

and Guo

J Q

2013 Journal of Glaciology and Geocryology 35 857 DOI:10.7522/j.issn.1000-0240.2013.0097

Using the precipitation data from 121 representative observation stations in Northwest China from 1961 to 2011, the characteristics of spring precipitation were analyzed. EOF, REOF and power spectrum method were used to research the spatial and temporal characteristics of spring precipitation in Northwest China. Whether there were mutations in the spring precipitation sequence was tested by Mann-Kendall test. The results show that the spatial distribution of spring precipitation in Northwest China was extremely uneven. One can see that the rainy areas located in the southeast and northwest, and the less rain areas located in the middle. Spring precipitation was consistent in the entire region at the first space scale, two natural climate zones could be recognized with the second space scale, and six natural climate zones could be recognized with the third space scale. The 1980s was the most precipitation decade among the five decades. The 1970s was least precipitation decade. The annual variability of spring precipitation was significant in Northwest China. Precipitation in most precipitation year was three times more than that in least precipitation year. There were obvious mutations in spring precipitation in Northwest China from 1961 to 2011. An abrupt decreasing occurred in 1973, and an abrupt increasing occurred in 1985. As regards of changing cycles, there was a long period of 18 19 years (the main) and short periods of 5 years and 7 years. It is expectancy that spring precipitation in Northwest China would be slowly coming down in the future 20 years.

1. Institute of Arid Meteorology, China Meteorological Administration/Key Laboratory of Arid Climatic Change and Disaster Reduction of Gansu Province/Key Laboratory of Arid Climate Change and Disaster Reduction of China Meteorological Administration, Lanzhou Gansu 730020, China; 2. Zhangye Meteorological Bureau, Zhangye Gansu 734000, China; 3. Meteorological Service Center of Gansu Province, Lanzhou Gansu 730020, China; 4. Northwest Regional Climate Center, Lanzhou Gansu 730020, China

利用西北地区121个气象站1961-2011年降水量资料, 分析了西北地区春季降水的基本气候特征;通过EOF、REOF、功率谱等方法, 对西北地区春季降水的时空特性进行了研究, 用Mann-Kendall检验法检验西北地区春季降水序列是否存在突变现象.结果表明: 西北地区春季降水空间分布极不均匀, 其空间分布特征是东南部和西北部为多雨区、中间为少雨区.西北地区春季降水在第一空间尺度上为全区一致, 在第二空间尺度上可分为2个自然气候区, 在第三空间尺度上可分为6个自然气候区.从年代际变化来看, 1980年代是近半个世纪来降水最多的10 a, 1970年代是降水最少的10 a;西北地区春季降水的年际变率十分显著, 降水最多的年份是最少年份的3倍多. 1961-2011年间西北地区春季降水发生了明显的突变: 1973年出现了一次趋于减少的突变, 1985年出现了一次趋于增多的突变. 18~19 a的长周期是其主要周期, 其次是5 a和7 a的短周期. 未来20 a西北地区春季降水量呈缓慢下降的趋势.

... ^[1] The frequent occurrences of drought have been threatening food production, social security, thereby creating tremendous pressure and challenges to the socio-economic conditions and environment-ecology problems ...

2004

0.0

1.301

Liu

H L

, Li

D L

and Guo

J Y

2004 Journal of Glaciology and Geocryology 26 55

Using EOF and REOF methods, precipitation data from May to July from 19 representative stations in Hexi corridor up to 2002 were analyzed to reveal the drought characteristics and abnormal space distribution of precipitation in Hexi corridor from the late spring to the early summer. Basing on climate sectoring, inter-decadal variability of precipitation is discussed for the first time series and every representative station, and the coefficient between the first time series and the prophase circulation factor is calculated on the 500 hPa geopotential height in northern hemisphere. It is found that precipitation in Hexi corridor in the late spring and early summer has a better coherence in the first space. There are three nature sectors in the second space. There are five nature sectors in the third space. Precipitation in the 1980s is the maximum of the recent 5 decades. Precipitation in the 1990s decreased, and from the end of the 20th century to the beginning of the 21st century precipitation has an obvious increasing trend in China. If the early Eurasian meridional circulation strengthens in winter, then polar vortex area will enlarge, polar vortex intension will strengthen and cold air will be more active, propitious to precipitation in Hexi corridor in the late spring and early summer. Wave train in Europe, Tibetan Plateau, North China and West Pacific results in a typical pluvial stream with high in east and low in west on the 500 hPa geopotential height, especially the ram and move of East Asian trough. Foreshadow model of precipitation in the late spring and early summer in representative station has a fitting percent of about 72.

1. Zhangye Meteorological Bureau, Zhangye Gansu 734000, China; 2. Cold And Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences, Lanzhou Gansu 730000, China; 3. Lanzhou Institute of Arid Meteorology, China Meteorological Administration, Lanzhou Gansu 730020, China

利用河西走廊19个气象代表站建站至2002年5～6月降水量资料,分析了河西走廊春末夏初干旱的基本气候特征;在利用EOF和REOF方法进行降水空间异常变化分析和气候分区的基础上,讨论了第一时间系数(PC1)及各区代表站降水量的年代际变化规律.结果表明,河西走廊春末夏初降水量在第一空间尺度上为全区一致;在第二空间尺度上可分为3个气候区;在第三空间尺度上可分为5个自然气候区.1980年代为近50a来降水最多的10a,1990年代有所减少,20世纪末至21世纪初有明显增加.前期冬季欧亚径向环流加强,亚洲区极涡面积扩大、强度加强,冷空气活动频繁,将有利于次年春末夏初河西走廊降水偏多.欧洲-青藏高原-华北-西太平洋的波列,特别是东亚大槽的填塞和青藏高原低值系统频繁活动,造成了500hPa高空场上"东高西低"的典型多雨流型.

... ^[2] For example, the 2003 summer#cod#x2013 ...

2007

0.0

1.371

... ^[3] the 2005/2006 autumn#cod#x2013 ...

2011

0.799

0.363

Y H

, Xu

H M

and Liu

2011 Acta Meteorologica Sinica 25 176 DOI:10.1007/s13351-011-0025-8

The spatial-temporal features of the extremely severe drought and the anomalous atmospheric circulation in summer 2006 are analyzed based on the NCEP/NCAR reanalysis data, the characteristic circulation indices given by the National Climate Center of China, and the daily precipitation data of 20 stations in the east of Southwest China (ESC) from 1959 to 2006. The results show that the rainless period started from early June and ended in early September 2006 with a total of more than 80 days, and the rainfall was especially scarce from around 25 July to 5 September 2006. Precipitation for each month was less than normal, and analysis of the precipitation indices shows that the summer precipitation in 2006 was the least since 1959. The extremely severe drought in the ESC in summer 2006 was closely related to the persistent anomalies of the atmospheric circulation in the same period, i.e., anomalies of mid-high latitude atmospheric circulation, western Pacific subtropical high (WPSH), westerlies, South Asian high, lower-level flow, water vapor transport, vertical motion, and so on. Droughts usually occur when the WPSH lies anomalously northward and westward, or anomalously weak and eastward. The extreme drought in summer 2006 was caused by the former. When the WPSH turned stronger and shifted to the north and west of its normal position, and the South Asian high was also strong and lay eastward, downdrafts prevailed over the ESC and suppressed the water vapor transfer toward this area. At the same time, the disposition of the westerlies and the mid-high latitude circulation disfavored the southward invasion of cold air, which jointly resulted in the extremely severe drought in the ESC in summer 2006. The weak heating over the Tibetan Plateau and vigorous convective activities over the Philippine area were likely responsible for the strong WPSH and its northwestward shift in summer 2006.

1.Key Laboratory of Meteorological Disaster of Ministry of Education, Nanjing University of Information Science & Technology, Nanjing, 210044 China 2.Chongqing Institute of Meteorological Sciences, Chongqing, 401147 China 3.Chongqing Municipal Meteorological Bureau, Chongqing, 401147 China

... ^[4,5] the 2007/2008 winter#cod#x2013 ...

2011

0.799

0.363

... ^[4,5] the 2007/2008 winter#cod#x2013 ...

2012

0.0

1.166

... ^{[6#cod#x2013 ...}

2011

0.0

1.948

Song

, Yang

and Li

C Y

2011 Chinese Journal of Atmospheric Sciences 35 1009 DOI:10.3878/j.issn.1006-9895.2011.06.02

To understand the mechanisms responsible for the severe drought in Yunnan Province during the boreal winter (DJF) of 2009/2010, by using NCEP/NCAR reanalysis data and winter precipitation index in Yunnan Province which is calculated from station precipitation data in Yunnan Province, this study investigates the relationship between the North Atlantic Oscillation (NAO) and the precipitation in Yunnan Province during boreal winters. In the 49 winters of 1961/1961－2009/2010 period, the correlation coefficient between the time series of winter mean precipitation indices in Yunnan Province and the NAO indices is 0.373 (exceeds the 95% confidence level), which indicates that the winter precipitation in Yunnan Province is linked to the variability of the NAO. Regression results between the winter precipitation indices in Yunnan Province and the northern hemispheric anomalous geopotential height at 300 hPa show that the circulation systems associated with the winter precipitation in Yunnan Province have a south and a north major components: the southern branch trough and the ridge of Lake Baikal. And the results show that the NAO can impact these two components through the quasi-stationary waves propagating along the Asia－Africa subtropical jet and the wave reflections, respectively. Therefore, the variations of the NAO and the winter precipitation in Yunnan Province are linked. The results also show that the linkage between the NAO and the winter precipitation in Yunnan Province is modulated by ENSO phenomenon. The relationship between the NAO and the winter precipitation in Yunnan Province is much close in the warm ENSO winters, while, they are barely linked in the cold ENSO winters.

为了揭示2009/2010年冬季云南出现严重干旱灾害的原因, 本文利用NCEP/NCAR 再分析资料以及云南省台站降水资料计算得到的云南冬季降水指数, 讨论了在北半球冬季 (12～2月) 北大西洋涛动 (North Atlantic Oscillation, 简称NAO) 和云南省降水 (旱涝) 之间的联系。分析结果表明, 在1961/1962～2009/2010年间的49个冬季, 冬季平均的云南降水指数和NAO指数的相关系数为0.373 (超过95%信度), 表明NAO和云南省冬季降水之间存在着一定的联系。云南省冬季降水指数和北半球300 hPa位势高度异常场之间的回归结果表明, 和云南冬季降水联系紧密的环流系统可分为南北两个主要部分: 南支槽波列和贝加尔湖高压脊系统。研究表明, NAO可以通过在亚—非副热带急流中传播的准定常波列以及波的反射分别影响这两个系统。因此, NAO和云南冬季降水之间存在一定的联系。同时, 我们的结果还表明NAO和云南冬季降水之间的关系还受ENSO事件的调制。在ENSO为暖位相的冬季, NAO和云南降水关系更为紧密, 而在ENSO为冷位相的冬季, NAO和云南冬季降水的联系则不太显著。

2010

0.0

... 8] He et al ...

2011

0.0

2.937

J Y

, Zhang

M J

, Wang

S J

and Wang

X M

2011 Acta Geographica Sinica 66 1179(in Chinese)

Based on the daily data of temperature and precipitation of 108 meteorological stations in Southwest China from 1960 to 2009, we calculated the surface humid indexes of months and years, as well as the extreme drought frequency. According to the data, the temporal and spatial characteristics of the extreme drought frequency in inter-annual, inter-decadal, summer monsoon and winter monsoon has been analyzed. The results are indicated as follows: (1) In general, the southwest of Sichuan Basin, the southern part of Hengduan Mountains, southern coast of Guangxi and north of Guizhou are the areas in which the extreme drought frequency has significantly increased in the past 50 years. For the decadal change, from the 1960s to the 1980s the extreme drought frequency has presented a decreasing trend, and the high frequency area appeared alternately in the southeast - northwest - east of Southwest China; the 1990s is the most wettest decade and it is wet over the whole area. In the 2000s, the extreme drought frequency rises quickly, but the regional differences reduce. (2) In summer monsoon, the extreme drought frequency is growing, which generally happens in the high mountains around the Sichuan Basin, most parts of Guangxi and "the broom-shaped mountains". It is obvious that the altitude has impacts on the extreme drought frequency of summer monsoon; in winter monsoon, the area is relatively wet and the extreme drought frequency is decreasing. (3) In summer monsoon, the abrupt change happened in 2003, while the abrupt change of winter monsoon happened in 1989, and the joint abrupt changes in different periods lead to the extreme drought frequency variation. The departure sequence vibration of annual extreme drought frequency is quasi 5 years and quasi 12 years.

Geography and Environment College of Northwest Normal University, Lanzhou 730070, China

利用中国气象局整编的1960-2009 年西南地区108 站逐日气温、降水等资料,计算年、月地表湿润指数,并进行标准化,统计极端干旱发生频率,对年际、年代际、季风期和非季风期的极端干旱变化特征进行分析,得出结论：(1) 整体上,四川盆地西南部、横断山区南端、广西南部沿海和贵州北部是近50 年来年极端干旱发生频率明显增加的地区;年代际变化上,20 世纪60-80 年代极端干旱呈逐渐减少趋势,高发区交替出现在东南—西北—东,90 年代下降明显,整个地区都转湿,进入21 世纪后,极端干旱距平呈现正距平,且增幅较大,区域间差异却显著减小。(2) 季风期与非季风期的极端干旱变化有很大差异,季风期极端干旱频率在不断增加,多发生在四川盆地周边海拔较高的山区、广西大部和“帚形山脉”地带,海拔对季风期极端干旱发生频率有一定影响;非季风期缓慢下降,整体偏湿。(3) 通过滑动t 检验和小波分析发现,季风期西南极端干旱在2003 年发生突变,非季风期在1989 年突变,年极端干旱发生频率是季风期和非季风期的突变叠加的结果;年极端干旱存在准5年和准12 年的周期变化。

... ^[9] reported characteristics of extreme drought climate change in Southwest China over nearly 50 years ...

2013

0.0

1.688

Zheng

J M

, Zhang

W C

, Wan

Y X

and Duan

2013 Plateau Meteorology 32 1665 DOI:10.7522/j.issn.1000-0534.2012.00143

Using the NCEP/NCAR reanalysis data from 1961 to 2010, the difference of atmospheric circulation in spring between the extreme drought and rainy years (from March to April) in Yunnan were analyzed. The results show that the characteristis of atmospheric circulation in rainy spring are different from that of in spring of drought year. In the extreme drought years, it is negative anomaly in the region of the north of 40°N but positive in the region of the south of 40°N on 500 hPa geopotential height field. The south branch trough and Siberian high pressure are weak in the extreme drought year. The atmospheric circulations are characteristic of strong westerly and zonal circulation on 700 hPa in the Eurasia mid-latitude. There is an abnormal anticyclone over the Bay of Bengal and the South Peninsula, which lead the sinking motion in the troposphere. In the rainy years, it is positive anomaly in the region of the north of 38°N but negative of the south of 38°N on 500 hPa geopotential height field. The atmospheric circulations are characteristic of weak westerly and meridional circulation on 700 hPa in the Eurasia mid-latitude. The south branch trough and Siberian high pressure are strong in the rainy years. There is an abnormal cyclone from north of South Peninsula to Yunnan, which lead the ascending motion in the troposphere. Water vapor transportation over Yunnan comes from the vapor transportation of south branch westerly and water vapor transportation of west Pacific subtropical high external vapor transportation. Both of the two water vapor transportation are weak, resulting in the reducing of the vapor content over Yunnan in the extreme drought years. On the other hand, the convergence of those two branch vapor transportations are strong, especially stronger in the of water vapor western Pacific subtropical high external, increasing the content of the water vapor in the rainy year. The weak cold air, the sinking air flow and the weak convergence of water vapor flux, which lead to the poor dynamical conditions and weak water vapor transportation, are the reasons of extreme drought in Yunnan. The AO index is negative correlation of the temperature in Yunnan in spring but positive correlation of the precipitation. In the negative phase of AO index, the temperature is higher and the precipitation is less in Yunnan, the spring drought is obviously, which is helpful to the development of extreme drought.

利用1961-2010年NCEP/NCAR全球逐月资料，对云南4次极端干旱年春季（3-4月）的大气环流特征与多雨年春季的大气环流特征进行了合成对比分析。结果表明，云南极端干旱年春季与多雨年春季的大气环流特征有明显的差异。干旱年， 500 hPa高度距平场上欧洲、亚洲以40°N为界北负南正，南支槽偏弱, 地面西伯利亚高压偏弱； 700 hPa中纬度为西风距平，西风带偏强\, 气流平直，以纬向环流为主，低纬地区孟加拉湾、中南半岛有异常反气旋环流，整个对流层为下沉运动。降水偏多年, 500 hPa高度距平场上欧亚中高纬西低东高，亚洲以38°N为界北正南负，南支槽偏强, 地面西伯利亚高压偏强；中纬度为东风距平，西风带偏弱，环流经向度较大，低纬地区中南半岛北部到云南为异常气旋环流，对流层中低层为上升运动。水汽通量的多年平均表明, 影响云南的水汽是南支西风水汽输送带和西太平洋副热带高压外围的水汽输送带在云南上空交汇。干旱年, 两支水汽输送带上的水汽通量辐合弱，云南上空水汽含量减少；降水偏多年, 两支水汽输送带上的水汽通量辐合强，特别是西太平洋副热带高压外围的水汽输送带辐合强，云南上空水汽含量增加。干旱年, 中高纬的冷空气难以影响到云南，低纬下沉气流和水汽通量辐合弱, 使云南产生降水的动力条件和水汽条件均不足; 而多雨年则相反，产生降水的动力条件和水汽条件均有利。同期AO指数与云南春季气温呈负相关, 与降水呈正相关\.AO为负位相时云南高温少雨，易出现春旱，有利于极端干旱的发生。

... ^[10] analyzed the spring abnormal circulation of extreme drought years in Yunnan ...

2010

0.799

0.363

... ^[11] reported the vapor transport of drought and flood in the summer in Southwest China ...

2012

0.799

0.363

... ^[12] studied the asymmetric relation between winter North Atlantic Oscillation and the precipitation in Southwest China ...

2010

0.0

1.688

Y H

, Xu

H M

, Bai

Y Y

and Li

2010 Plateau Meteorology 29 523(in Chinese)

利用1959\_2006年西南地区东部20个测站和中国740个测站的逐日降水量资料，采用EOF、 Mann-Kendal、 MESA（最大熵谱分析）、 Morlet小波、线性相关等方法, 分析了西南地区东部夏季降水的时空演变特征。结果表明：西南地区东部夏季降水量大体分为3种类型，即全区域一致型、南北相反型和东西相反型，其中全区域一致型是西南地区东部夏季降水最主要的分布型，出现频率达到60%以上。在整个时间域内，西南地区东部夏季降水略有增加的趋势；在20世纪60年代和70年代总体偏旱，而在80年代前期总体偏涝， 80年代中后期开始波动加剧，旱涝交替发生，但总体相对偏涝; 21世纪初以来，西南地区东部总体偏旱； 1959年以来， 1998, 1980和1993年是降水偏多最明显的年份，而2006年和1972年降水偏少最明显。西南地区东部夏季降水的年际及年代际变化特征均较明显，存在2～3年、 15年左右的显著周期。西南地区东部与宜宾以下的整个长江流域和西藏东部及川西高原的降水呈显著正相关，而与华南及东南沿海地区、华北中部和四川盆地西部地区呈显著负相关。

... ^[13] reported spatial and temporal evolution characteristics of summer precipitation in the east of Southwest China ...

2014

0.0

1.948

... ^[14] analyzed the main mode of precipitation in dry or wet season in Southwest China ...

2013

0.0

... ^[15,16] ...

2012

1.148

1.2429

... ^[15,16] ...

2007

0.0

... ^[17] The Mann#cod#x2013 ...

... Kendall testing method is used to determine the temporal locations of abrupt changes of spring precipitation,^[17] with a priori set a #cod#x003D ...

... ^[17] A t-testing is used to test the statistical significance of abrupt changes ...

... ^[17] ...

... ^[17] The dominant timescales of variability are 15 years and 25 years ...

... ^[17] The Mann#cod#x2013 ...

... ,^[17] for a given time series of n samples ...

2008

1.148

1.2429

Zhang

and Wan

S Q

2008 Chin. Phys. B 17 2311 DOI:10.1088/1674-1056/17/6/063

Based on physical backgrounds, the four time series of the Guliya(Tibetan plateau) ice core (GIC) \textit{$\delta $}$^{18}$O, andthree natural factors, i.e. the rotation rate of earth, sunspots, andEl Nino--Southern Oscillation (ENSO) signals, are decomposed into twohierarchies, i.e. more and less than 10-year hierarchiesrespectively, and then the running $t$-test is used to reanalyse thedata before and after filtering with the purpose of investigating thecontribution of natural factors to the abrupt climate changes in thelast one hundred years. The results show that the GIC \textit{$\delta$}$^{18}$O evolved with a quasi-period of 7--9 years, and the abruptclimate changes in the early 1960s and in the period from the end ofthe 1970s to the beginning of the 1980s resulted from the jointeffect of the two hierarchies, in other words, the two interdecadalabrupt changes of climate in the last one hundred years were global.The interannual variations of ENSO and sunspots were the importanttriggering factors for the abrupt climate changes in the last onehundred years. At the same time, the method of Information Transfer(IT) is employed to estimate the contributions of ENSO signals andsunspots activities to the abrupt climate changes, and it is foundthat the contribution of the interannual variation of ENSO signals isrelatively large.

a Laboratory for Climate Studies of China MeteorologicalAdministration, National Climate Center, Beijing 100081, China;Yangzhou Meterorological Office, Yangzhou 225000,China; b Tongda College, Nanjing University of Post andTelecommunications, Nanjing 210003, China;Key Laboratory of Regional Climate--Environment Research forTemperate East Asia, Institute of Atmospheric Physics, ChineseAcademy of Sciences, Beijing 100029, Chi

... ^{[18#cod#x2013 ...}

2005

0.811

0.4541

2008

1.148

1.2429

Zhang

D Q

, Feng

G L

and Hu

J G

2008 Chin. Phys. B 17 736 DOI:10.1088/1674-1056/17/2/062

Using the daily precipitation data of 740 stations in China from1960 to 2000, the analysis on the variations and distributions ofthe frequency and the percentage of extreme precipitation to theannual rainfall have been performed in this paper. Results indicatethat the percentage of heavy rains (above 25mm/day) in the annualrainfall has increased, while on average the day number of heavyrains has slightly reduced during the past 40 years. In the end of1970s and the beginning of 1980s, both the number of days withextreme precipitation and the percentage of extreme precipitationabruptly changed over China, especially in the northern China. Bymoving t test, the abrupt change year of extreme precipitation foreach station and its spatial distribution over the whole country arealso obtained. The abrupt change years concentrated in 1978--1982for most regions of northern China while occurred at variousstations in southern China in greatly different/diverse years.Besides the abrupt change years of extreme precipitation at partstations of Northwest China happened about 5 years later incomparison with that of the country's average.

a College of Physics Science and Technology, Yangzhou University, Yangzhou 225009, China; b College of Physics Science and Technology, Yangzhou University, Yangzhou 225009, China;Laboratory for Climate Studies of China Meteorological Administration, National Climate Center, Beijing 100081,China; c Laboratory for Climate Studies of China Meteorological Administration, National Climate Center, Beijing 100081,China;The Institute of Atmospheric Physics, Chinese Academy of Sciences, Beijing 100029, China

... 20] ...

2004

0.0

1.166

... ^[21] predicts an increasing precipitation in the next eight decades ...

... ,^[21] is season-dependent ...