Interdecadal change in the relationship between the Antarctic oscillation and autumn rainfall in the Yunnan–Guizhou plateau of Southwest China

Abstract

The interdecadal change in the relationship between the Antarctic Oscillation (AAO) and autumn rainfall in the Yunnan–Guizhou Plateau of Southwest China is investigated by using the observed autumn rainfall data at 119 stations and the National Centers for Environment Prediction/National Center for Atmospheric Research reanalysis data for the period 1961–2021. Results show the AAO correlates well with the autumn rainfall in the Yunnan–Guizhou Plateau for the second period (2002–2021) because the AAO becomes stronger. The possible influencing mechanism of AAO on autumn rainfall in the Yunnan–Guizhou Plateau during 2002–2021 is related to the meridional teleconnection pattern and associated convection over the Philippine Sea. The positive AAO can trigger a meridional teleconnection pattern in the upper troposphere to propagate from the southern Pacific to northern Pacific and cause anomalous westerly over the tropical west Pacific, which inhibits the convection over the Philippine Sea. On the one hand, the weakened convection over the Philippine Sea causes the anomalous ascending motion over the Yunnan–Guizhou Plateau; on the other hand, it results in an anomalous anticyclone over the tropical Northwest Pacific and strengthens the transport of water vapor from the tropical Pacific to the Yunnan–Guizhou Plateau.

INTRODUCTION

A significant warming trend has been observed in most regions of the world over the past several decades, and meanwhile the annual rainfall has obviously increased in most parts of the land (IPCC AR5, 2014, 2021; Marvel et al., 2019). Under the background of global warming, the climate in most areas of southern China has shown a trend of warming and wetting, but it has shown a trend of warming and drying in Southwest China (Dai, 2021; Gu & Adler, 2015; Shi et al., 2014). The rainfall in Southwest China is affected by the Asian monsoon circulation and the complex terrain. Thus, its spatial and temporal distribution is uneven, which can be divided into the dry season and wet season (Lu et al., 2020; Wang et al., 2012). The large interannual variability of rainfall causes the frequent occurrences of droughts and floods in Southwest China, which have significant impacts on local production and the social economy (Huang et al., 2015; Zhang et al., 2013; Zhao et al., 2018).

The Yunnan–Guizhou Plateau, located in southwest China, is one of the four largest plateaus in China, where the rainfall is mainly concentrated in summer and autumn, accounting for more than 80% of the annual rainfall (Lu et al., 2020; Wan et al., 2016). Many studies have shown that abnormal rainfall is well related to anomalous atmospheric circulations. For example, when the East Asian summer monsoon is enhanced and the South Asian summer monsoon is weakened, the western Pacific subtropical high is weakened, and the anomalous northeasterly prevails over southern China, which inhibits the northward transport of water vapor from the tropical oceans and causes the decrease of summer rainfall (Cao et al., 2012, 2016; Chao & Chen, 2001; Huan & Li, 2018). The enhanced low latitude westerly can strengthen the transport of water vapor from the Indian Ocean and the South China Sea, which contributes to the increased summer and autumn rainfall (Yan et al., 2012; Yuan & Yang, 2020; Zhang et al., 2011). The zonal shift of the cyclone (anticyclone) circulation in the equatorial northwestern Pacific also contributes to the autumn precipitation anomaly in the Yunnan–Guizhou Plateau (Xu et al., 2016; Zhang et al., 2013).

The atmospheric teleconnection pattern plays an important role in modulating the regional rainfall change (Linderholm et al., 2011; Wallace & Gutzler, 1981; Wang et al., 2017). Some studies have reported that the negative Arctic Oscillation (AO) can weaken the transport of water vapor from the Oceans, which causes a decrease in the autumn rainfall in the Yunnan–Guizhou Plateau (Barriopedro et al., 2012; Kong et al., 2012; Yang et al., 2012). The Antarctic Oscillation (AAO) is one of the most prominent atmospheric circulation modes in the Southern Hemisphere, which has an important impact on the atmospheric circulation in both the southern and northern hemispheres (Baldwin, 2001; Fan & Wang, 2007; Gong & Wang, 1999; Nan & Li, 2003; Song & Li, 2009; Thompson & Wallace, 2000). The AAO also can influence the precipitation and air temperature in China (including Southwest China) via the meridional teleconnection pattern and air-sea interaction (Fan & Wang, 2004; Nan et al., 2009; Qian, 2014; Zhang et al., 2017, 2024). Recent studies suggest that the relationship between the precipitation over Southwest China and tropical sea surface temperatures exhibits remarkable interdecadal change (Ke et al., 2021; Liu et al., 2018). Then, does the relationship between the AAO and the autumn rainfall in the Yunnan–Guizhou Plateau also exhibit obvious interdecadal change? What are the possible influencing mechanisms? This scientific question is the focus of the present study.

STUDY DATA AND METHOD

The monthly rainfall data at 119 stations were extracted from the China Meteorological Data Network (Figure 1a). The autumn rainfall can account for 20%–30% of the total annual rainfall in the Yunnan–Guizhou Plateau (Figure 1b). The atmospheric reanalysis data supplied by the National Center for Environment Prediction (NCEP)/National Center for Atmospheric Research (NCAR) (Kalnay et al., 1996). The data are divided into 17 layers, including horizontal wind, vertical velocity, and specific humidity. The Outgoing Longwave Radiation (OLR) data provided by the European Centre for Medium-Range Weather Forecast (ECMWF) Reanalysis V5 (ERA5). The indices of the Niño 3.4 are all from the Climate Prediction Center of the National Oceanic and Atmospheric Administration. The time lengths of all the selected datasets cover the period of 1961–2021 and summer refers to the boreal autumn months (September, October and November).

[IMAGE OMITTED. SEE PDF]

In this study, we select the modified Antarctic Oscillation Index (AAOI), which is defined by the standardized difference of zonal mean sea level pressure (SLP) between 40° S and 70° S because the negative correlation in the zonal mean SLP anomalies between 40° S and 70° S is stronger than that between 40° S and 65° S (Gong & Wang, 1999; Nan & Li, 2003). The autumn precipitation index (API) in the Yunnan-Guizhou Plateau is derived from the normalized autumn rainfall regionally averaged over the 119 stations during 1961–2021.

RESULTS

Many studies have paid more attention to revealing the anomalous characteristics of summer rainfall associated with atmospheric circulations in the Yunnan–Guizhou Plateau (Cao et al., 2016; Yuan & Yang, 2020), and few studies focus on the autumn rainfall, which also has an important contribution to the annual total rainfall (Figure 1b). So only the relationship between the AAO and the rainfall in autumn are discussed in this study. To further investigate the relationship between the AAO and rainfall in autumn, the normalized time series of AAOI and API in autumn is shown in Figure 2a. In the early period (e.g., before the early 2000s), the API does not show coherent variability and the AAOI with a correlation coefficient of 0.03, and a significant in-phase relationship between the AAOI and API only appears after the early-2000s with a correlation coefficient of 0.61 over 95% confidence level. Figure 2b shows the interdecadal change in the AAOI-API relationship, and it is clear that this relationship has enhanced remarkably after the early 2000s. So, the total analysis period of 1961–2021 is divided into two periods: 1961–2001 and 2002–2021. As seen by Figure 2c, the surface level pressure becomes weaker at 70° S and stronger at 40° S, especially in the Pacific, which enhances the difference of surface level pressure between 40° S and 70° S. So the AAO becomes stronger during 2002–2021 than that during 1961–2001.

[IMAGE OMITTED. SEE PDF]

Some previous studies pointed out that the AAO could excite teleconnection patterns from the southern hemisphere to the northern hemisphere, which was a possible way for AAO to affect the climate in the northern hemisphere (Fan & Wang, 2006, 2007; Qian, 2014). Figure 3a shows the regression of zonal wind at 200 hPa against the AAOI during 1961–2001. It is clear that the AAO can trigger an obvious meridional teleconnection pattern in the upper troposphere over the Pacific. The positive AAO can cause the anomalous westerly over the equatorial Western Pacific and easterly over the North Indian Ocean. Meanwhile, the significant meridional teleconnection pattern with quasi-barotropic structures propagates from the southern Pacific to the northern Pacific, especially in the mid-upper troposphere (Figure 3c). From the South Pole to the North Pole, the anomalous easterly alternates with the anomalous westerly. Figure 3b shows the regression of zonal wind at 200 hPa against the AAOI during 2002-2021. The meridional teleconnection pattern in the upper troposphere triggered by the AAO becomes more significant over the Pacific, and the range of anomalous westerly influenced by the AAO covers the whole tropical western Pacific. In the same way, the meridional teleconnection pattern also has a quasi-barotropic structure, and from the South Pole to the North Pole, the anomalous easterly alternates with the anomalous westerly (Figure 3d). The positive AAO can result in significant anomalous westerly over tropical western Pacific, which is different from that in Figure 3c.

[IMAGE OMITTED. SEE PDF]

How does the meridional teleconnection pattern affect precipitation in China? The convection over the tropical oceans plays an important role in linking the AAO and precipitation in China (Qian, 2014). Because of the important effect of the ENSO on the interannual variability of the AAO (Carvalho et al., 2005; Zhou & Yu, 2004), to avoid the link between the AAO and the convection over the tropical oceans investigated in this study is a “spurious” link due to the fact that both are affected by ENSO events, we use the partial correlation to remove the effects of the Niño 3.4. Figure 4a shows the partial correlation between the AAOI and the OLR after removing the Niño 3.4 index during 1961–2001. It is clear that the AAOI negatively correlates well with the OLR, especially in Australia and the Bay of Bengal, which indicates that the convection is strengthened. During 2002–2021 (Figure 4b), the distribution of partial correlation between the AAOI and the OLR is different from that in Figure 4a. The AAOI positively correlates well with the OLR over the tropical Northwest Pacific, which indicates that the convection is weakened.

[IMAGE OMITTED. SEE PDF]

Some previous studies suggested that the convections over the Western Maritime Continent (10° S–5° N, 90–115° E) and the Philippine Sea (10–20° N, 120–140° E) were well related to the summer and autumn rainfall in Southwest China (Jiang et al., 2017; Nitta & Hu, 1996; Qian, 2014; Wang et al., 2015). Then, which convection can be linked to the relation of the AAO and autumn rainfall in the Yunnan–Guizhou Plateau? Figure 5 displays the interannual variation of averaged OLR over the Western Maritime Continent (Figure 5a) and Philippine Sea (Figure 5b) and the API during 2002–2021. After removing the effect of Niño 3.4 index, the partial correlation coefficients are 0.09 and 0.49, respectively, which confirms that the convection over the Philippine Sea plays an important role in linking the AAO and autumn rainfall, well matched to the Figure 4b.

[IMAGE OMITTED. SEE PDF]

Figure 6a shows the latitude-height cross-section of partial correlation between the OLR averaged over the Philippine Sea (10–20° N, 120–140° E) and the autumn vertical velocity, meridional wind and vertical velocity vector along 98–110° E during 2002–2021 after removing the effect of Niño 3.4 index. It is clear that the convection over the Philippine Sea is well related to the vertical motion over the Yunnan–Guizhou Plateau. When the convection over the Philippine Sea becomes weaker, the anomalous ascending motion and southerly will control the Yunnan–Guizhou Plateau, which is favorable to forming more autumn rainfall. Meanwhile, the negative anomalous convection over the Philippine Sea corresponds to an anomalous anticyclone over the tropical Northwest Pacific (Figure 6b), and it can transport more water vapor form the tropical ocean to the Yunnan–Guizhou Plateau, which provides favorable water vapor conditions for the occurrence of autumn rainfall.

[IMAGE OMITTED. SEE PDF]

CONCLUSION AND DISCUSSION

In summary, the relationship between the AAO and autumn rainfall in the Yunnan-Guizhou Plateau has shown an obvious interdecadal change over the past several decades. For the first period (1961–2001), the AAO is not well related to autumn rainfall and the correlation coefficient only is 0.03. For the second period (2002–2021), the AAO correlates well with the autumn rainfall, and the correlation coefficient can reach 0.61, which can pass the 95% confidence level. The AAO becomes stronger for the second period than that for the first period, which plays an important role in modulating the autumn rainfall in the Yunnan–Guizhou Plateau.

Why does the relationship between the AAO and autumn rainfall in the Yunnan–Guizhou Plateau exhibit remarkable interdecadal change? The meridional teleconnection pattern and associated convection over the tropical Northwest Pacific play important roles in linking the AAO and autumn rainfall in Yunnan–Guizhou. Both during 1961–2001 and 2002–2021, the AAO can trigger meridional teleconnection patterns with quasi-barotropic structures over the Pacific from the southern hemisphere to the northern hemisphere in the upper troposphere, which can cause the anomalous westerly over tropical west Pacific. From 2002 to 2021, the significant anomalous westerly covers more of the range of tropical west Pacific, including the Philippine Sea, which weakens the climatology of easterly. So, the convection over the Philippine Sea is inhibited. Furthermore, the weakened convection over the Philippine Sea can cause anomalous ascending motion over the Yunnan–Guizhou Plateau; meanwhile, an anomalous anticyclone corresponding to the weakened convection can strengthen the transport of water vapor from the tropical west Pacific. Both of the above contribute to the positive autumn rainfall anomaly in the Yunnan-Guizhou Plateau from 2002 to 2021.

In the current study, we have revealed the interdecadal change in the relationship between the AAO and autumn rainfall in the Yunnan-Guizhou Plateau, but what causes the interdecadal change is still unclear. The AAO has experienced an obvious interdecadal change from negative anomalies to positive anomalies in recent decades and has an enhanced impact on precipitation and atmospheric circulation in China (Bian & Lin, 2009; Sun et al., 2013). We also found that the AAO became stronger during 2002–2021 than that during 1961–2002. Then why does the AAO become stronger in recent years? Some studies suggest the sea ice in the South Pole can influence the AAO (Wang et al., 2022; Wu & Zhang, 2011). In the past several decades, the extent of sea ice at the South Pole has shown an increasing trend (Hobbs et al., 2016; Holland, 2014). Does this contribute to the strengthened AAO? It needs to be considered in future studies.

AUTHOR CONTRIBUTIONS

Yusen Li: Methodology; software; data curation; writing – original draft; investigation. Yong Zhao: Conceptualization; investigation; funding acquisition; writing – original draft. Lixia Meng: Investigation; writing – original draft.

ACKNOWLEDGMENTS

We are grateful to NCEP/NCAR and ECMWF for releasing the reanalysis data. This study is sponsored by The Natural Science Foundation of Yunnan Province (202302AN360006) and Scientific Research Foundation of CUIT (Grant No. KYTD202333). The authors appreciate the comments and suggestions from Dr. Y. J. Li and X. Zhou during this study. We are very grateful to two anonymous reviewers for the very constructive comments during the review process.

CONFLICT OF INTEREST STATEMENT

The authors declare no conflict of interest.

DATA AVAILABILITY STATEMENT

The authors declare that the main data supporting the findings of this study are available from the following websites: Observational rainfall data from . NCEP/NCAR Reanalysis dataset from and Niño 3.4 time series from . ERA5 Reanalysis dataset from .

References

Baldwin, M.P. (2001) Annular modes in global daily surface pressure. Geophysical Research Letters, 28(21), 4115–4118. Available from: [DOI: https://dx.doi.org/10.1029/2001GL013564]