---

La Niña and its Impacts on China's Climate

Wang Shao-wu
Peking University
Beijing, China
swwang@pku.edu.cn

Introduction

The impacts of El Niño events on the climate have been widely studied from view point of statistics and dynamics. However, the counterpart of the El Niño -- La Niña -- has been examined much less than El Niño in China. It has been found that the number of typhoon landings in China are greatly reduced in an El Niño year, and that Northeast China usually suffers with cool summer which significantly diminishes crop yields.

However, divergence occurred when the impact of El Niños on the summer precipitation changes in China was examined. Identifying the response of the summer precipitation to ENSO events, we had to do the classifications of El Niño events according to the timing or geographical location of the positive anomaly. Of course, introducing any kind of classification in to the analysis increases the level of subjectivity and adds difficulty in applying climatic predictions. The lower significance of the relationship between ENSO events and precipitation in China may relate not only to the characteristics of ENSO events, but also to other factors such as atmospheric circulation in the middle latitudes and sea surface temperatures of the warm pool in the Western Pacific. The latter has exhibited behavior independent to ENSO, although it negatively correlated with the SSTs in the Eastern Pacific. In this discussion paper no stratification approach was adopted in examining the impacts of ENSO events on the climate of China. The impacts of La Niña events have been emphasized.

Identification of ENSO events

Usually, ENSO events have been identified according to a single series of SOI or SST in the Eastern Equatorical Pacific, e.g., the SST in Niño3. However, uncertainty of any series increases in the earlier part of the record (such as in the late 19th century or early 20th century). Thus, it is desirable to use a composite index instead of a single one. Wang and Shi (1992) indicated that correlation coefficient was 0.65 between seasonal anomalies of SOI and SST in Niño C (0-10°S, 90-180°W) for the period 1871-1989. It infers that each of them can explain only less than one half of the variance of the others; a composite index will carry more information than a single index. Therefore, a composite ENSO index series was constructed on the basis of following four indices:

1. Niño 3 SST (Kaplan et al., 1997, Cane et al., 1997). Spring 1856 to Autumn 1991. Since 1991, the data set was updated by use of SST data from NOAA's CPC.
2. Niño C SST (Angell, 1981). Summer 1867 to Winter 1987. Since 1987 and gaps in the time series in the early period were updated using the data of Wang and Shi (1992).
3. SOI. Allan (1991); Ropelewski and Jones (1987). Spring 1866 to Fall 1997. Since 1997, data of NOAA's CPC were used.
4. SOI (Shi and Wang, 1989). The methodology used in construction of this series was the same as CPC suggested, but stational data of SLP in the early years were different from those of Ropelewski and Janes (1987).

First of all, each of the four series was normalized itself for whole four seasons together in reference to the period of 1961-1990. Then, the four normalized series were averaged to find a composite ENSO index series, for the period spring of 1867 to winter of 1997. Before 1867, series of four indices were not complete. In the making of the composite ENSO index series, positive SST anomalies were averaged with negative SOI anomalies. Then, a positive composite index corresponds to positive phase of SST and negative phase of SOI, and vice versa. Finally, A-events (negative composite index) for the period of 1867-1997 were identified, when the composite index was -0.5 and persisted at least two seasons.

Impacts of La Niña Events

32 A-events were found for the period of 1867-1997, and listed in Table l. The seasons of Spring, Summer, Fall and Winter were denoted by 1, 2, 3, and 4, respectively. Impacts of La Niña on the climate were investigated by using the following six climatic indices:

I1. Number of typhoon landfalls

I2. Summer temperature anomalies averaged for ten stations covering East Asia between 40°-53°N.

I3. Summer precipitation anomalies averaged for five stations in north and west China (Beijing, Zhengzhou, Yichang, Chengdu, and Chongqing).

I4. Summer precipitation anomalies averaged for five stations in south China (Fuzhou, Shantou, Guangzhou, Zhanjiang, and Nanning).

I5. Winter precipitation anomalies averaged for five stations in northwest China (Taiyuan, Zhengzhou, Xian, Yinchuan, and Lanzhou).

I6. Winter precipitation anomalies averaged for five stations in southeast China (Wenzhou, Fuzhou, Jian, Guangzhou, and Guilin).

Table l indicates that precipitations increased in the north and west, and decreased in the south in summer, and increased in the northwest and decreased in the southeast in winter when A-events occurred. Meanwhile, landing typhoons increased, and summer temperatures over East Asia were higher than normal. However, the latter two indices correlated weaker than the others to ENSO. Correlation coefficients calculated by using the whole time series for 1880-1997 are given in the last line. Figures with a star show that the correlation is significant at the 99% confidence level.

Table 1

La Niña	I-1	I-2	I-3	I-4	I-5	I-6
1880					/	/
1882					*	*
1886			*		*
1887	*		*	*
1889			*
1890			*		*
1892	*				*
1893	*		*	*	*	*
1894	/	/	/	/	*	*
1898					*
1903			*		*	*
1906			*	*		*
1908
1909	*		*	*		*
1910	*			*		*
1916		*		*		*
1917		*	*			*
1921			*			*
1922			*			*
1924	*	*
1933	*	*		*	*	*
1937	/	/	/	/
1938		*	*	*		*
1947			*		/	/
1950		*	*		*	*
1954			*	*	*	*
1955		*	*
1956			*	*	/	/
1964				*	/	/
1970		*		*
1971	*				*
1973		*	*			*
1974	*				/	/
1975		*		*	*	*
1988				*	*
F	9/33	10/33	18/33	14/33	14/29	17/29
r	-0.10	-0.25	-0.36	0.26*	-0.23*	0.48*

La Niña Summit Home Page | Table of Contents | ESIG | Summary | Executive Summary