Asymmetric relationships between El Ni?o/La Ni?a and floods/droughts in the following summer over Chongqing,China

XIANG Bo,ZHOU Jie and LI Yonghua

Chongqing Climate Center,Chongqing,China

ABSTRACT This study presents a detailed analysis of the asymmetric relationships between the warm/cold phase of the El Ni?o–Southern Oscillation and the typical flood/drought years in summer over Chongqing.Furthermore,its underpinning mechanisms are also explored.The results show that:(1)El Ni?o and La Ni?a have an asymmetric influence on summer precipitation in the following year over Chongqing. Generally, the composite atmospheric circulation anomalies for El Ni?o years are consistent with the composite results for typical flood years in summer over Chongqing,which indicates a tight link between typical flood years in summer over Chongqing and El Ni?o events.However,the relationship between typical drought years in summer over Chongqing and La Ni?a events is not significant. (2) From winter to the following summer, the extent of positive SST anomalies in the equatorial eastern Pacific associated with typical flood years in summer over Chongqing shrinks,whereas in the tropical Indian Ocean,the extent slightly expands.This trend indicates that the impact of El Ni?o on typical flood years in summer over Chongqing is maintained through the‘relay effect’of SSTs in the tropical Indian Ocean,which is the result of a lagged response of positive SST anomalies in the tropical Indian Ocean to El Ni?o events in winter.

KEYWORDS El Ni?o/La Ni?a;asymmetry;Chongqing;typical drought and flood years

1. Introduction

The El Ni?o–Southern Oscillation(ENSO)is the strongest signal found in the interannual variability of the tropical ocean–atmosphere system. ENSO directly affects the weather and climate in the tropical Pacific,while it indirectly affects the weather and climate in East Asia through ‘teleconnections’ with atmospheric circulation patterns(Zhang,Sumi,and Kimoto 1999;Zhai et al.2016;Chen et al.2018).It has been shown that,in addition to the asymmetry in their inherent amplitude, structure and temporal evolution (Hoerling, Kumar, and Zhong 1996;Wu,Li,and Zhou 2010),El Ni?o and La Ni?a exhibit an asymmetry in their impacts on the East Asian climate(Zhang,Sumi,and Kimoto 1996;Zhang et al.2015;Yuan,Li,and Yang 2014;Chen et al.2017).Ni et al.(1995)found that El Ni?o and La Ni?a have asymmetric impacts on the East Asian winter monsoon — the East Asian winter monsoon is substantially weakened during the El Ni?o period,whereas it is not considerably strengthened during the La Ni?a period.Zhang,Sumi,and Kimoto(1996)and Chen(2002)also came to the same conclusion.Such asymmetric impacts are mainly due to the asymmetry of the anomalous anticyclone and anomalous cyclone that occur over the Northwest Pacific during the winter halfyear of El Ni?o and La Ni?a, respectively (Zhang et al.2015;Li,Zhang,and Wen 2017).During the winter halfyear of El Ni?o, a significant anomalous anticyclone occurs over the Northwest Pacific, which results in an enhancement of water vapor transport towards the south and thus causes a considerable increase in precipitation over the southeast coast and South China(either winter or summer or total annual precipitation);during the winter half-year of La Ni?a, the cyclonic circulation anomalies over the Northwest Pacific are markedly weakened,which does not substantially reduce the water vapor transport from the south, thus leading to a non-significant negative precipitation anomaly over South China (Zhang, Min, and Su 2017; Li, Zhang, and Wen 2017).

Moreover, Xu, Feng, and Chen (2016) found that El Ni?o and La Ni?a also show a remarkable asymmetry in the conversion relationship between the East Asian winter monsoon and the following summer monsoon.During the El Ni?o year, an anomalous anticyclonic circulation over the Northwest Pacific is maintained from winter to the following summer, whereas the southerly winds in the west weaken the East Asian winter monsoon and strengthen the summer monsoon in the following year (Chen, Feng, and Wu 2013); during the La Ni?a year,an anomalous cyclonic circulation over the Northwest Pacific is maintained from winter to the following summer, which results in weaker summer monsoon in the following year (Chen et al. 2017). However,the anomalous anticyclonic circulation over the Northwest Pacific has a more southerly location and higher intensity than the anomalous cyclonic circulation.Therefore, El Ni?o exerts a greater impact on the East Asian summer monsoon and the following summer monsoon compared to La Ni?a (Xu, Feng, and Chen 2016). The maintenance of the anomalous Northwest Pacific anticyclone from winter to the following summer is mainly achieved through increased sea surface temperatures(SSTs)in the tropical Indian Ocean during the summer following El Ni?o(Annamalai,Liu,and Xie 2005;Yuan et al. 2010; Chen et al. 2016). Xie et al. (2009)indicated that in the summer following the peak of El Ni?o, the significant positive SST anomaly (SSTA) over the tropical Indian Ocean induces the Ekman divergence mechanism over the Northwest Pacific by stimulating equatorial atmospheric Kelvin waves, which in turn leads to the formation of the anomalous anticyclone.

Despite extensive research regarding the asymmetry of the impacts of El Ni?o and La Ni?a on circulation and precipitation, few studies of summer droughts and floods over Chongqing have been carried out.Chongqing is located in the eastern part of Southwest China,with distinct climatic characteristics compared to other parts of the southwest region (Li et al. 2010). In recent years,there has been an upward trend in both the intensity and frequency of drought and flood disasters over Chongqing. For example, regional or local catastrophic floods occurred in the summers of 1998, 2004,current study, the causes of typical summer droughts and floods in Chongqing are analyzed from the perspective of the asymmetry in the impacts of El Ni?o and La Ni?a on droughts and floods in the following summer over Chongqing. The study will provide evidence for predicting summer droughts and floods over Chongqing.

2. Data and methods

The data used in this study include(1)monthly precipitation data at 34 national meteorological stations(Figure S1)throughout Chongqing,provided by the Chongqing Meteorological Information Centre; (2) global monthly sea surface pressure,height,wind(zonal and meridional components),and specific humidity NCEP–NCAR reanalysis data(Kalnay et al.1996),with a spatial resolution of 2.5° × 2.5° in the horizontal direction and 17 isobaric surfaces from 1000 to 10 hPa in the vertical direction;and (3) monthly SST data released by NOAA, with a spatial resolution of 2° × 2° (Huang et al. 2017). All data are selected for the period 1961–2017, and the climate state is the 30-year average from 1981 to 2010.

To reflect the overall situation of monthly summer precipitation (droughts/floods) over Chongqing,a regional precipitation index (γ, %) for the period 1961–2017 is calculated using the method adopted by the National Climate Center of the China Meteorological Administration(Zhao 1999):

where n is the number of stations,Riis the total precipitation for June–August,is the average annual precipitation,i is the sequence number of a station(i=1,2,...,n),and n+is the number of stations with a precipitation anomaly ΔR ≥0 among the n stations. The index γ can depict the level of regional precipitation well; a higher γ value indicates more regional precipitation(Li et al.2012).

3. Results

3.1 Influence of El Ni?o/La Ni?a on summer precipitation in Chongqing

2007, and 2014 (Ding and Hu 2003; Chen and Ding 2006),whereas regional catastrophic or severe droughts occurred in the summers of 2006 and 2011 (Li, Xu, and Liu 2009; Li, Chen, and Dong 2014). These drought and flood disasters have caused extremely serious economic losses and major social impacts across this region.In the

It’s well known that ENSO is an important factor influencing the weather and climate in East Asia, the relationships between the warm/cold phase of ENSO and the typical flood/drought years in summer over Chongqing is further analyzed in this section.

We use composite analysis to analyze the influence of El Ni?o/La Ni?a on summer precipitation in Chongqing.Using the criterion of one standard deviation of the Ni?o3 SST index,we select eight El Ni?o years(1972,1982,1986,1991,1997,2002,2009,and 2015)and eight La Ni?a years(1967,1970,1973,1975,1988,1999,2007,and 2010).The atmospheric circulation at various levels is composited for the summer following the abovementioned El Ni?o and La Ni?a years (Figure 1). Compared with the composite results for typical drought and flood years in summer over Chongqing(Figures S2–S6),the composited anomalies of the atmospheric circulation at various levels for El Ni?o years exhibit similar characteristics to the composite results for typical folod years. In the wind field at 850 hPa,there exists an abnormal anticyclonic circulation near the western Pacifci, and the entire region of Chongqing is under the control of anomalous southwesterly winds, with easterly anomalies over the equatorial western Pacifci to the Indochina Peninsula(Figure 1(c)).In the height feild at 500 hPa,a positive geopotential height anomaly area is found over the Ural Mountains and Lake Baikal, with stronger high-pressure ridges. Meanwhile,there are significant geopotential height anomalies over the tropical zone,whereas the western Pacific subtropical high is intensified and shifts westward and southward(Figure 1(b)).At 200 hPa,the East Asian subtropical westerly jet is stronger and moves southward (Figure 1(a)),whereas the anomalous characteristics of the East Asian subtropical westerly jet are more evident for typical flood years in summer over Chongqing(Figure S5(a)).The consistency demonstrates that El Ni?o is an important external forcing leading to typical flood years in summer over Chongqing.

Figure 1. Composite (a, d) 200-hPa zonal wind anomalies, (b, e) 500-hPa geopotential height anomalies, and (c, f) 850-hPa wind anomalies in summer for the(a–c)El Ni?o and(d–f)La Ni?a years.Shaded areas represent values reaching the 95%confidence level.

However,the composite results for atmospheric circulation in La Ni?a years show substantial differences compared with the those for typical drought years in summer over Chongqing. At 850 hPa, for the La Ni?a years, the southwesterly anomalies that control Chongqing are obviously stronger, which extend northward to Inner Mongolia;there are also clear easterly anomalies over the equatorial eastern Pacific (Figure 1(f)). In contrast, for the typical drought years, the southwesterly anomalies are weaker and accompanied by significant westerly anomalies over the equatorial eastern Pacific(Figure S5(f)).At 500 hPa,for the La Ni?a years, the high-latitude region represents a significant negative anomaly area, whereas in the midlatitude region, the height field is close to the climatic average (Figure 1(e)). For the typical drought years, the high-latitude region represents a significant positive anomaly area, whereas the midlatitude region is a significant negative anomaly area (Figure S5(e)). Moreover, in the tropical ocean, the equatorial western Pacific represents a significant negative anomaly area over a broad area for the La Ni?a years (Figure 1(e)), whereas for the typical drought years only the South China Sea is a significant negative anomaly area (Figure S5(e)). At 200 hPa, there are no characteristics of a northward shift in the position of the East Asian subtropical westerly jet or a westward shift in the center of the jet for the La Ni?a years (Figure 1(d)).

Similarly, this asymmetry is also displayed in the SST field.During the preceding winter of typical flood years in summer over Chongqing, the equatorial eastern Pacific shows an obvious El Ni?o condition;the equatorial western Pacific SST is slightly colder (Figure 2(b)),which is highly consistent with the distribution characteristics of SSTAs in winter for the El Ni?o years (figure not shown). In the preceding winter, the SSTA field composited for typical drought years in summer of Chongqing has not reached the intensity of La Ni?a events and thus lacks the characteristics of La Ni?a(Figure 2(c)), although there is a weak negative SSTA in the equatorial eastern Pacific.

Based on the above analysis,the typical flood years in summer of Chongqing are tightly linked to El Ni?o events,which further shows that El Ni?o is an important external forcing leading to more precipitation in summer over Chongqing. However, the typical drought years in summer of Chongqing are not significantly related to La Ni?a,which to some extent indicates that the impact of La Ni?a on summer precipitation in Chongqing may not be completely opposite to the impact of El Ni?o.

3.2 Possible mechanisms underpinning the impacts of El ni?o on typical flood years in the following summer in Chongqing

El Ni?o events generally peak in autumn and winter,whereas the decay of El Ni?o begins in the following spring. During the following summer, El Ni?o tends to disappear, with relatively weak SSTAs in the equatorial central and eastern Pacific. How does El Ni?o, which peaks around the end of the year, exert a lagged effect on summer precipitation in the following year? Earlier studies have shown that during autumn and winter of the El Ni?o development year,the East Asian subtropical region stimulates an anomalous anticyclonic circulation over the Northwest Pacific through Rossby wave teleconnection (Zhang, Sumi, and Kimoto 1999; Wang, Wu,and Fu 2000; Wang and Zhang 2002). However, during the summer of the El Ni?o decay year,the maintenance of the Northwest Pacific anomalous anticyclone is mainly the result of a lagged response of SSTAs in the tropical Indian Ocean (Xie, Hafner, and Tokinaga 2009; Chen,Yuan, and Yang 2013; Huang et al. 2016; Yuan, Gao,and Liu 2017;Chen et al.2019).Therefore,is the impact of El Ni?o on typical flood years in the following summer over Chongqing also a result of the lagged response of SSTAs in the tropical Indian Ocean?

To answer this question, the correlation coefficients between the regional precipitation index and SST in different seasons are calculated (Figure 3). For the preceding winter,the significant positive correlation area is mainly located in the equatorial eastern Pacific,whereas the negative correlation area is located in the western North Pacific (Figure 3(a)). Additionally, the tropical Indian Ocean appears to be a positive correlation area.For spring,the extent of the significant positive correlation area in the equatorial eastern Pacific clearly shrinks,whereas in the tropical Indian Ocean the extent of the significant positive correlation area substantially expands (Figure 3(b)). For summer, the extent of the significant positive correlation area in the equatorial eastern Pacific is further reduced,whereas in the tropical Indian Ocean it further expands,especially in the northern Indian Ocean and the South China Sea(Figure 3(c)).From the preceding winter to the summer,the range of the significant positive correlation area in the equatorial eastern Pacific shrinks, whereas in the tropical Indian Ocean the range slightly expands. These changes are consistent with the composite results (figure not shown),which further verify that the typical flood years in summer over Chongqing are the result of a lagged response of positive SSTAs in the tropical Indian Ocean.

4. Conclusions and discussion

In this study,the asymmetric relationship between typical flood/drought years in summer over Chongqing and El Ni?o/La Ni?a in the preceding winter is analyzed in detail using NCEP–NCAR atmospheric circulation data, NOAA SST data, and precipitation data from 34 stations in Chongqing.Furthermore,the mechanisms underpinning the impacts of El Ni?o/La Ni?a are explored. The main conclusions are as follows:

Figure 2. Composite SSTAs in the preceding winter in (a) typical flood years and (b) typical drought years. Dotted areas represent values reaching the 90%confidence level.

(1) El Ni?o and La Ni?a have an asymmetric relationship with summer precipitation in the following year over Chongqing. Generally, the composite results for the atmospheric circulation anomalies in El Ni?o years are consistent with those for typical flood years in summer over Chongqing, which indicates a tight link between typical flood years in summer over Chongqing and El Ni?o events. However, the relationship between typical drought years in summer over Chongqing and La Ni?a events is not significant.

(2) From winter to the following summer,the extent of positive SSTAs in the equatorial eastern Pacific associated with typical flood years in summer over Chongqing shrinks, whereas in the tropical Indian Ocean, the extent slightly expands. This trend indicates that the impact of El Ni?o on typical flood years in summer over Chongqing is maintained through the ‘relay effect’ of SSTs in the tropical Indian Ocean, which is the result of a lagged response of positive SSTAs in the tropical Indian Ocean to El Ni?o events in winter.

Figure 3.Distribution of the correlation between the regional precipitation index and SST in the(a)preceding winter,(b)spring,and(c)summer.Dark shaded areas represent values reaching the 95%confidence level.

This study has analyzed the asymmetry in the impacts of El Ni?o and La Ni?a on summer precipitation in the following year over Chongqing. The results indicate a close relationship between typical flood years in summer over Chongqing and El Ni?o events,whereas the relationship between typical drought years in summer over Chongqing and La Ni?a events is not significant. How, therefore, does La Ni?a affect summer precipitation in Chongqing? For the summer following La Ni?a, the reduction in precipitation over Chongqing is not large as expected and exhibits more complex characteristics. Among the eight La Ni?a years selected in this study, only three years exhibit less precipitation, whereas the remaining five years have more precipitation. This phenomenon merits further attention in our future work.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

This research was financially supported by the General Project of Technical Innovation and Application Demonstration in Chongqing, China [grant number cstc2018jscx-msybX0165],the Special Fund for the Development of Key Technology in Weather Forecasting of the China Meteorological Administration [grant number YBGJXM (2018)04-08], and the National Natural Science Foundation of China [grant number 41875111].