Cite this Article
SHI Wen-Jing, XIAO Zi-Niu. Impact of the Preceding Boreal Winter Southern Annular Mode on the Summertime Somali Jet. Atmospheric and Oceanic Science Letters, 2014, 7(6): 534-539  
Permissions
Copyright?2014, Editorial office of Atmospheric and Oceanic Science Letters
This is an Open Access article under the terms of CCAL.
Impact of the Preceding Boreal Winter Southern Annular Mode on the Summertime Somali Jet
SHI Wen-Jing1,2, XIAO Zi-Niu1,2,3,*
1. Department of Atmospheric Science, Nanjing University of Information Science & Technology, Nanjing 210044, China
2. China Metrological Administration Training Center, Beijing 100081, China
3. Institute of Atmosphere Physics, Chinese Academy of Sciences, Beijing 100029, China
*Corresponding author: XIAO Zi-Niu,xiaozn@cma.gov.cn
Abstract

One of the major high-latitude circulation systems in the Southern Hemisphere is the Southern Annular Mode (SAM). Its effect on the Somali Jet (SMJ), which connects the Southern and Northern hemispheres, cannot be ignored. The present reported results show that time series of both the Southern Annular Mode Index (SAMI) during the preceding winter and the summertime Somali Jet intensity Index (SMJI) display a significant increasing trend and have similar interdecadal variation. The latter was rather strong around 1960, then became weaker up to the mid-1980s, before starting to strengthen again. The lead-lag correlations of monthly mean SAMI with the following summertime SMJI showed significant positive correlations in November, December, and January. There are thus connections across two seasons between the SAM and the SMJ. The influence of the winter SAM on the summer SMJ was explored via analyses of SST anomalies in the Southern Indian Ocean. During strong (weak) SAM/SMJ years, the SST east of Madagascar is colder (warmer) while the SST west of Australia is warmer (colder), corresponding to the positive (negative) Southern Indian Ocean Dipole-like (SIODL) event. Subsequently, the SIODL excites an anticyclone located over the Arabian Sea in summer through air-sea coupling from winter to summer, which causes an increase in the summer SMJ intensity. The anticyclone/high branch of the SAM over the Southern Hemisphere subtropics and the cyclone/low over the east coast of Madagascar play an important role in the formation of Southern Indian Ocean "bridge" from winter to summer.

Keyword: Southern Annular Mode; Somali Jet; Southern Indian Ocean Dipole; ocean-atmosphere coupled bridge
1 Introduction

In the early 1920s, the relationship between southern hemispheric circulation and low-level cross-equatorial flows (CEFs) were studied by Simpson (1921), who first noticed that CEF is located along the east coast of Somalia (i.e., the Somali jet; SMJ) and is connected to low pressure systems over the Asian continent and high pressure systems over the southern hemispheric subtropical region. Subsequently, interactions between the southern and northern hemispheres were widely studied ( Li, 1936; Tao et al., 1962; Sun, 2010). In particular, Li (1936) discovered that cold air from the North Pole region can sweep across and down to the Equator, reaching the southern hemisphere and leading to precipitation and a temperature drop in that region. At the same time, the outbreak of cold air coming from Australasia can lead to the generation of typhoons in the Northwest Pacific when the cold air reaches the lower-latitude regions of the Northern Hemisphere. Tao et al. (1962) found that when the low westerly index circulation (high westerly index corresponds to zonal circulation; low westerly index corresponds to meridional circulation) prevails over the subtropical and tropical regions of eastern Asia, mass transport occurs from the Southern Hemisphere to the Northern Hemisphere. Unfortunately, a lack of data at high latitudes in the Southern Hemisphere meant that most early studies focused on the subtropical areas of the southern hemispheric atmospheric circulation, including the Mascarene High and the Australian High ( Yang and Huang, 1989; Xu et al., 1993). When studying the climatic characteristics of CEFs in summer, Shi et al. (2001) revealed that the South African High, the South Indian Ocean High, and the Australian High have a strong influence on the SMJ. He and Chen (1989) analyzed the influence of the quasi-40-day low-frequency oscillation at midlatitudes in the Southern Hemisphere on the Asian summer monsoon, noting that the cold air in the Southern Hemisphere first strengthens the Mascarene High, the Australian High, and subsequently the westerly flow to the south of those two highs through meridional propagation of the low-frequency oscillation, which then enhances the CEFs.

It was only in the mid-1980s that the impact of the Southern Annular Mode (SAM) on CEFs sparked interest and subsequent studies. The SAM, which is also called the Antarctic Oscillation ( Gong and Wang, 1998), is a large zonally symmetric and barotropic mode of variability of the extratropical circulation in the Southern Hemisphere. In particular, a strong-phase (weak-phase) SAM is characterized by negative (positive) sea level pressure anomalies over Antarctica and positive (negative) anomalies over the Southern Hemisphere midlatitudes ( Thompson and Wallace, 2000). Recent studies examined the influence of the SAM on the Asian monsoon climate (e.g., Sun et al., 2009; Fan and Wang, 2004; Fan, 2006), as well as the relationship between SAM anomalies and SMJ variability ( Gao et al., 2013; Qiu et al., 2014). As pointed out by Gao et al. (2013), in stronger SAM years the Mascarene High and the Australian High (especially in March) are stronger due to (a) the "see-saw" structure of the atmospheric circulation over the subtropics and the higher latitudes in the Southern Hemisphere, and (b) the deeper intertropical convergence zone (ITCZ). Thus, the pressure gradient between the subtropical and tropical regions tends to increase in the spring, meaning that the SMJ occurs earlier and stronger. The study of Gao et al. (2013) illustrated that the SAM in boreal winter can be used as a predictor of the onset of the SMJ in early spring. Besides, the SMJ is strongest in boreal summer and its influence on the Asian summer monsoon is also strongest in summer ( Lin et al., 2008; Lu and Lin, 2009; Shi and Xiao, 2013). Based on the above results, we raise the following questions to be answered in the present paper: (1) Can the influence of the SAM on the SMJ during boreal winter last from spring to summer? (2) What are the underlying physical mechanisms of this influence?

2 Datasets and method

The data employed in this study included: (1) monthly reanalysis data from the National Centers for Environmental Prediction-National Center for Atmospheric Research (NCEP-NCAR), with a spatial resolution of 2.5° ´ 2.5° from 1951 to 2010; (2) monthly sea surface temperature (SST) data (°C) from the National Oceanic and Atmospheric Administration (NOAA) Extended Reconstructed SST 2° ´ 2° dataset for the period 1951-2010. Correlation and composite statistical methods were used. Based on the numerical order of the SAM index (SAMI) over the nearly 60-year period, the years 1959, 1960, 1962, 1963, 1970, 1982, 1986, 1989, 1996, 1999, 2000, and 2002 were selected as years with positive/strong SAM anomalies. Similarly, based on the numerical order of the SMJ intensity index (SMJI), the years 1952, 1953, 1957, 1969, 1972, 1975, 1977, 1983, 1992, 1993, 2001, and 2006 were selected as years with negative/weak SAM anomalies. In addition, 1957, 1958, 1959, 1960, 1961, 1990, 1998, 2000, 2001, 2003, 2008, and 2010 were selected as years with strong SMJ anomalies, and 1966, 1968, 1969, 1972, 1973, 1974, 1975, 1977, 1979, 1981, 1995, and 1997 were selected as years with weak SMJ anomalies.

The SMJI was defined as the spatial mean of the meridional wind speed averaged from the surface to 700 hPa over SMJ regions (40-55°E at the equator). The SAMI was defined as the difference in the normalized monthly zonal mean sea level pressure (SLP) between 40°S and 70°S (Nan and Li, 2003; available online at http://ljp.lasg.ac.cn/ dct/page/ 65609).

3 Results
3.1 Relationships between SAM and SMJ

Figures 1a and 1b show the lead-lag correlations of the monthly mean between SAMI and SMJI during summertime (June, July, and August (JJA)). For the raw series (Fig. 1a), we found significant positive correlations for all months except for September and March. In particular, correlations for December, January, April, May, and June exceeded the 99% confidence level. For the detrended series (Fig. 1b), the correlation coefficient between the monthly mean SAMI and summertime SMJI was greatly reduced, and only the months of November, December, and January showed a significantly positive correlation. Therefore, the focus of the discussion in this paper is on the relationship between the SAM during the preceding winter (November, December, and January (NDJ)) and the summertime SMJ.

Figure 1 (a) Lead-lag correlations between summer SMJI and SAMI in different months; (b) same as for (a) but for detrended series; (c) standardized time series of SAMI in November, December, and January (NDJ) and SMJI in June, July, and August (JJA), and the correlation coefficient between them; (d) same as for (c) but for detrended series. The three dashed lines in (a) and (b) respectively indicate statistical at the 90%, 95%, and 99% confidence levels. The pink dashed line with open squares and the green solid line with hollow circles in (c) and (d) respectively represent the SAMI and SMJI series.

From the standardized time series of SAMI in NDJ and SMJI in JJA (Fig. 1c), we found that both the SAMI and SMJI yearly time series displayed a linear increasing trend with significance at the 99% confidence level. Furthermore, these two series possessed consistent interdecadal variations, both becoming stronger around 1960 and after the mid-1980s, and showing a weaker trend during the intervening period. The correlation coefficient between the two parameters was 0.463 (significant at the 99% confidence level). Besides, considering the influence of the strong positive trend in these two indices, the correlation of the detrended indices was then calculated. The correlation decreased, but the coefficient with a value of 0.28 was still significant at the 95% confidence level (Fig. 1d). Thus, it can be concluded that a close relationship exists between the boreal winter SAM and the following summertime SMJ.

To better illustrate the relationship between the boreal winter SAM and the following summertime SMJ, Figs. 2a and 2b show the composite differences of the boreal winter SLP in the Southern Hemisphere between strong and weak SMJ years. As confirmed by the composite analysis, the preceding winter circumpolar low deepens and the midlatitude high strengthens during strong summer SMJ years, and vice versa. The patterns were similar for the raw data (not shown), which indicates that the correlation between the preceding winter SAM and the summertime SMJ exists regardless of whether or not the linear trend is removed. Employing the same method, the summer wind anomalies difference between positive and negative preceding winter SAM years at 850 hPa were computed (Figs. 2c and 2d). In both the raw (not shown) and detrended data, the significant southerly wind anomalies over the SMJ areas occurred in strong SAM years, and northerly wind anomalies occurred in weak SAM years in summer.

Figure 2 Composite analyses of SLP (hPa) during the preceding winter NDJ in 12 (a) strong and (b) weak summer SMJ years. Also shown are composite analyses of summer 850 hPa wind anomalies (m s-1) in 12 (c) strong and (d) weak winter SAMI years. The linear trend has been removed. The dotted areas and black bold arrows denote composite differences at the 95% confidence level.

3.2 Correlations of Indian Ocean SST with SAM and SMJ

The above analysis reveals connections across two seasons between the SAM and SMJ. In this section, we explore the possible physical mechanism involved. We know that the ocean has a "memory", in the sense that it may store and then release climatic signals. So, it is necessary to consider the role of the ocean as the underlying surface. Figures 3a-c show composite maps of the winter, spring, and summer Southern Indian Ocean SST anomaly between strong and weak SAM years, respectively. During the strong (weak) preceding winter SAM years, strong cold (warm) SST anomalies develop east of Madagascar, while strong warm (cold) SST anomalies develop west of Australia. This SST anomaly dipole on the subtropical Indian Ocean can persist from the preceding boreal winter to boreal summer. This pattern is similar to the Subtropical Dipole Pattern (SDP), also known as the Southern Indian Ocean Dipole (SIOD), which was first described by Behera and Yamagata (2001), who addressed independence of this dipole from the tropical Indian Ocean Dipole (IOD). However, the positions of this pattern's dipole center are different from the SDP; therefore, we refer to it as the SIOD-like mode (SIODL). For clarification, the positive (negative) SIODL event is characterized by the occurrence of a warm (cold) SST anomaly in the southeastern (southwestern) Indian Ocean.

The SST patterns in Figs. 3d-f are similar to those in Figs. 3a-c, with a cold (warm) SST anomaly in the southeastern Indian Ocean and a warm (cold) SST anomaly in the southwestern Indian Ocean for strong (weak) SMJ years. The connection between the SIODL mode and the SMJ can also persist from winter to summer. According to the aforementioned analysis, we conclude that the SIODL mode is a "bridge" that connects the SAM to the SMJ. To further verify the relationship between the SIODL mode and the SMJ, an index (herein referred to as the SIODL mode index, SIODI) was obtained from the SST anomaly difference between the southeastern (30- 20°S, 50-75°E) and southwestern (20-5°S, 75-100°E) Indian Ocean. These regions correspond to areas of significant composite difference. The correlation coefficients between winter SIODI and spring SIODI, winter SIODI and summer SIODI, as well as spring SIODI and summer SIODI were 0.56, 0.28, and 0.6, respectively, meaning that the SIODL can persist from the preceding winter to summer, as demonstrated by Xu and Fan (2012, 2014) and Yang and Ding (2006). The winter, spring, and summer SIODI series, and the summer SMJI series were all consistent (not shown). The correlation coefficients between summer SMJI and winter, spring, and summer SIODI were 0.29, 0.24, and 0.30, respectively, at the 95% confidence level (the degrees of freedom: Ndof = 58). This result shows that stronger SMJ is usually related to a positive SIOD event ( Yang and Ding, 2007). Based on the above analysis, we conclude that the correlation between the preceding winter SAM and the summer SMJ is associated with the SIOD.

Figure 3 Composite differences of Indian Ocean SST anomalies (°C) during boreal (a) winter, (b) spring, and (c) summer between positive-phase and negative-phase preceding winter SAM years. Panels (d-f) are the same as (a-c) but for summer SMJ. The linear trend has been removed. The dotted areas denote composite differences at the 90% confidence level.

3.3 The "bridge" role of the Southern Indian Ocean

As mentioned above, the winter SAM can impact the following summer's SMJ through the SIODL, a manifestation of the "air-sea coupled bridge". To explain the atmospheric dynamics linking the SAM to the SMJ through the SIODL, we performed a correlation analysis between the SIODI and the atmospheric circulation (horizontal wind and geopotential height) at 850 hPa from winter to summer. As shown in Fig. 4, the SIODL is closely related to atmospheric circulation in the Southern Hemisphere. In the preceding winter (Fig. 4a), with a positive-phase SIODL, the zonal wind is enhanced over the southern hemispheric high latitudes and depressed over the southern hemispheric subtropics, co-varying with a deepening of the Antarctic circumpolar low and a strengthening of the subtropical high over the Southern Indian Ocean. Such a reverse variation in atmospheric circulations between the high and midlatitudes of the Southern Hemisphere is a major feature of the SAM and is consistent with our above analyses. It is also important to consider the cyclone/low located over the east coast of Madagascar. In spring (Fig. 4b), this cyclone strengthens gradually and starts to expand to the tropics and the southeast. On the other hand, the anticyclone/high branch of the SAM is greatly weakened and only the western part remains,strengthening the western Mascarene High. The positive- phase SAM is broken and enhanced southerly anomalies start to emerge over the Somali regions. As shown in Fig. 4c, the cyclone/low over the east coast of Madagascar continues to grow and the equatorial depression is deepened. Meanwhile, the anticyclone/high branch of the SAM nearly vanishes. Instead, the Antarctic circumpolar low is controlled by the large-scale anticyclone/high in the subtropical region, implying that the negative-phase SAM has formed. The anticyclone over the Arabian Sea and the increased pressure gradient between the tropical and high-latitude regions cause widespread southerly SST anomalies near the equator and, therefore, an increase in the summer SMJ intensity. Yang and Ding (2006) and Xu and Fan (2012, 2014), whose result is in accordance with the above analysis, also proposed that the SIOD influences variations in the SMJ through the local circulation anomaly excited by thermal forcing.

Figure 4 Coexistent correlation maps between the SIODI and wind vectors, as well as geopotential height (contour) anomalies at 850 hPa in (a) winter, (b) spring, and (c) summer after removal of the linear trend. The shaded areas ranging from light to dark represent the 90%, 95%, and 99% confidence levels, respectively. The black bold arrows denote composite differences at the 95% confidence level.

In addition, the large-scale cyclone located over the west of Australia (Fig. 4c) is related to the formation of the SIODL mode. Behera and Yamagata (2001) reported that the large-scale subtropical high in the Southern Hemisphere plays a role in the formation of the SDP. The subtropical cyclonic anomaly around 40°S contributes to the tropical easterly anomalies and the midlatitude westerly anomalies. These anomalies further result in latent heat flux anomalies, driving the Ekman heat transport and therefore upwelling. In turn, this upwelling changes the SST.

4 Conclusions

Statistical results indicate a significant influence of the southern hemispheric atmospheric circulation on the SMJ. The preceding winter SAMI and summertime SMJI series both displayed a linear increasing trend. Their interdecadal variations are also highly consistent, being weaker before the mid-1980s and stronger around 1960 and after the mid-1980s. The monthly lead-lag correlations between the monthly mean SAMI and the SMJI in JJA (for the raw series) showed significant positive correlations for all months except September and March. After removing the linear trend, the relationship between the monthly mean SAMI and the SMJI in JJA was much weaker and significant correlations were only observed in November, December, and January, which indicates a close relationship between the boreal winter SAM and the following summer's SMJ. A composite map of the summer SMJI with the SLP in the preceding winter showed that, during strong SMJ years, the southern circumpolar low of the Antarctic deepens and the subtropical-high band strengthens, and vice versa. This point was confirmed by a composite map of the winter SAMI with the summer wind anomaly at 850 hPa. Overall, the results showed that the preceding winter's SAM can provide a significant prophase signal for forecasting the summer SMJ.

Subsequent analyses verified that the influence of the winter SAM on the summer SMJ is likely a consequence of SST anomalies in the Southern Indian Ocean. In the SAM/SMJ strong (weak) years, the SST east of Madagascar is colder (warmer) while the SST west of Australia is warmer (colder), corresponding to a positive (negative) SIODL event. Moreover, these linkages can persist from the preceding winter to the summer. This result suggests that the winter SAM can impact the following summer SMJ through the SIODL mode, which acts as an "air-sea coupled bridge".

Generally, the anticyclone/high branch of the SAM over the southern hemispheric subtropics and the cyclone/low over the east coast of Madagascar play an important role in the formation of the Southern Indian Ocean "bridge". The strength of the cyclone/low over the east coast of Madagascar and the gradual disappearance of the anticyclone/high branch of the SAM from the preceding winter to the summer cause (a) the anticyclone to be positioned over the Arabian Sea, and (b) the pressure gradient between the tropical and high-latitude regions to strengthen, resulting in an increase in the summer SMJ intensity.

Acknowledgments. This study was jointly supported by the National Natural Science Foundation of China (41175051 and 41101045) and Plans to Graduate Research and Innovation Projects of Jiangsu Province Colleges and Universities (CXZZ13_0517).

Reference
1 Behera S. K. , T. Yamagata, 2001: Subtropical SST dipole events in the southern Indian Ocean, Geophys. Res. Lett. , 28(2), 327-330.
2 Fan K. , 2006: Atmospheric circulation anomalies in the Southern Hemisphere and summer rainfall over Yangze River Valley, Chinese J. Geophys. , 49(3), 672-679.
3 Fan K. , H. Wang, 2004: Antarctic oscillation and the dust weather frequency in North China, Geophys. Res. Lett. , 31, L10201, doi: DOI:10.1029/2004GL019465
4 Gao H. , Y. Liu, Y. Wang, et al. , 2013: Precursory influence of the Antarctic Oscillation on the onset of Asian summer monsoon, Chinese Sci. Bull. , 58, 678-683, doi: DOI:10.1007/s11434-012-5455-x.
5 Gong D. , S. Wang, 1998: Antarctic oscillation: Concept and applications, Chinese Sci. Bull. , 43(9), 734-738.
6 He J. , L. Chen, 1989: Quasi-40-day oscillation at mid-latitude of Southern Hemisphere with their relation to the summer monsoon over the Northern Hemisphere, Trans. Atmos. Sci. (in Chinese), 12(1), 11-18.
7 Li X. , 1936: Researches on East Asian cold invasion, in: Chinese Modern Science Treatise Collections-Meteorology (1919-1949), Science Press, Beijing, 35-118.
8 Lin M. , K. Fan, H. Wang, 2008: Somali jet changes under the global warming, J. Meteor. Res. , 22, 502-510.
9 Lu R. , Z. Lin, 2009: Role of subtropical precipitation anomalies in maintaining the summertime meridional teleconnection over the western North Pacific and East Asia, J. Climate, 22, 2058-2072.
10 Nan S. , J. Li, 2003: The relationship between the summer precipitation in the Yangtze River valley and the boreal spring Southern Hemisphere annular mode, Geophys. Res. Lett. , 30(24), 2266, doi: DOI:10.1029/2003GL018381.
11 Qiu J. , Z. Sun, W. Deng, 2014: The interdecadal variations of the vertical structure of the summer Somali cross-equatorial flow, Acta Meteor. Sinica (in Chinese), 72, doi: DOI:10.11676/qxxb2014.005.
12 Simpson G. C. , 1921: The South-West monsoon, Quart. J. Roy. Meteor. Soc. , 47(199), 151-171.
13 Shi N. , D. Shi, M. Yan, 2001: The effects of cross-equatorial current on South China Sea monsoon onset and drought/flood in East China, Acta Meteor. Sinica (in Chinese), 17(4), 405-414.
14 Shi W. , Z. Xiao, 2013: Variation of the cross-equatorial moisture transport in Somali and its impact on china early summer rainfall in nearly 60 years, Meteor. Mon. (in Chinese), 39, 39-45.
15 Sun J. , 2010: Possible impact of the boreal spring Antarctic Oscillation on the north American summer monsoon, Atmos. Oceanic Sci. Lett. , 3(4), 232-236.
16 Sun J. , H. Wang, Y. Wei, 2009: A possible mechanism for the co-variability of the boreal spring Antarctic Oscillation and the Yangze River valley summer rainfall, Int. J. Climatol. , 29, 1276-1284.
17 Tao S. , S. Xu, Q. Yun, 1962: The characteristics of the zonal and meridional circulation over tropical and subtropical regions in Eastern Asian in summer, J. Meteor. Res. (in Chinese), 32(2), 91-103.
18 Thompson D. W. J. , J. M. Wallace, 2000: Annular modes in the extratropical circulation. Part I: Month-to-month variability, J. Climate, 13, 1000-1016.
19 Xu X. , T. Zhao, J. He, 1993: The effect of Australian continent thermal forcing on anomalous variations of circulation systems in the Northern and Southern Hemispheres, Chinese J. Atmos. Sci. (in Chinese), 17(6), 641-650.
20 Xu Z. , K. Fan, 2012: Possible process for influences of winter and spring Indian Ocean SST anomalies interannual variability mode on summer rainfall over eastern China, Chinese J. Atmos. Sci. , 36(5), 876-888.
21 Xu Z. , K. Fan, 2014: Simulating the mechanism of the interannual variability mode of the Indian Ocean sea surface temperature anomalies on the summer rainfall over eastern China, Climatic Environ. Res. , 19(1), 31-40.
22 Yang M. , Y. Ding, 2006: Diagnostic study of the variation of Indian Ocean sea surface temperature and its impact on Indian summer monsoon rainfall, Acta Oceanol. Sinica (in Chinese), 28(4), 9-16.
23 Yang M. , Y. Ding, 2007: A study of the impact of South Indian Ocean Dipole on the summer rainfall in China, Chinese J. Atmos. Sci. (in Chinese), 31(4), 685-694.
24 Yang X. , S. Huang, 1989: The influence of intensity change of Mascarene High on the general circulation of atmosphere-a numerical experiment, J. Meteor. Sci. , 9(2), 125-138.