The Role of Barrier Layer in Southeastern Arabian Sea During the Development of Positive Indian Ocean Dipole Events

Transcription

1 J. Ocean Univ. China (Oceanic and Coastal Sea Research) DOI /s ISSN , (2): The Role of Barrier Layer in Southeastern Arabian Sea During the Developent of Positive Indian Ocean Dipole Events GUO Feiyan 1), LIU Qinyu 1), *, ZHENG Xiao-Tong 1), and SUN Shan 2) 1) Physical Oceanography Laboratory and Key Laboratory of Ocean-Atosphere Interaction and Cliate in Universities of Shandong, Ocean University of China, Qingdao , P. R. China 2) NOAA Earth Syste Research Laboratory, 325 Broadway, Boulder, Colorado 80305, USA (Received October 11, 2012; revised Noveber 26, 2012; accepted February 5, 2013) Ocean University of China, Science Press and Springer-Verlag Berlin Heidelberg 2013 Abstract Using data fro Argo and siple ocean data assiilation (SODA), the role of the barrier layer (BL) in the southeastern Arabian Sea (SEAS: 60 E 75 E, 0 10 N) is investigated during the developent of positive Indian Ocean Dipole (IOD) events fro 1960 to It is found that warer sea surface teperature (SST) in the northern Indian Ocean appears in June in the SEAS. This war SST accopanying anoalous southeastern wind persists for six onths and a thicker BL and a corresponding thinner ixed layer in the SEAS contribute to the SST waring during the IOD foration period. The excessive precipitation during this period helps to for a thicker BL and a thinner ixed layer, resulting in a higher SST in the SEAS. War SST in the SEAS and cold SST to the southeast of the SEAS intensify the southeasterly anoaly in the tropical Indian Ocean, which transports ore oisture to the SEAS, and then induces ore precipitation there. The ocean-atosphere interaction process aong wind, precipitation, BL and SST is very iportant for the anoalous waring in the SEAS during the developent of positive IOD events. Key words sea surface teperature (SST); ixed layer; barrier layer; Indian Ocean Dipole (IOD); persistence; precipitation; southeastern Arabian Sea 1 Introduction The variability associated with sea surface teperature (SST), wind and precipitation in the tropical Indian Ocean has been identified and defined as the Indian Ocean Dipole (IDO) ode (Saji et al., 1999; Webster et al., 1999; Murtugudde et al., 2000; Yaagata et al., 2003; Luo et al., 2010; Zheng et al., 2010). The Indian Ocean Dipole ode displays an anoalous east-west SST gradient and is accopanied with wind and precipitation anoalies in the tropical Indian Ocean, so it is also referred to as the Indian Ocean zonal ode (Le Blanc and Boulanger, 2001; Huang and Kinter, 2002). During Indian Ocean Dipole events, the changes in SST are found to be closely associated with changes in surface wind, the equatorial wind reverses direction fro westerlies to easterlies, and changes in sea surface wind are associated with a basinwide anoalous Walker circulation (Yaagata et al., 2002). According to previous studies, the Bjerknes-type (Bjerknes, 1969) feedback (Saji et al., 1999; Li et al., 2003), the wind-evaporation-sst feedback (Wang et al., 2003; Li et al., 2003; Halkides et al., 2006) and the oceanic Rossby waves (Webster et al., 1999; Xie et al., 2002; * Corresponding author. Tel: E-ail: liuqy@ouc.edu.cn Huang and Kinter, 2002) are doinant echaniss for the developent of IOD events. During an IOD event, the SST in the Indian Ocean experiences large changes. Saji et al. (1999) analyzed six extree events to present the life cycle of a typical dipole ode event. According to their studies, the northern Indian Ocean gets unusually war in the initial developent of a positive IOD event. Previous studies ainly discussed the SST cooling in the tropical southeastern Indian Ocean and the SST waring in the equatorial southwestern Indian Ocean during an IOD foration. So far it is still not very clear why the war SST appears in the northwest Indian Ocean like the one in the southeastern Arabian Sea (SEAS: 60 E 75 E, 0 10 N), why there is a corresponding southeasterly wind anoaly during the entire foration process of an IOD event, and whether there are other physical echaniss contributing to the SST anoaly in the SEAS in addition to the windtherocline-sst feedback, wind-evaporation-sst feedback, and ocean Rossby wave feedback echaniss. The interediate layer between the botto of ixed layer (ML) and the top of the therocline is called the barrier layer (BL) (Godfrey and Lindstro, 1989; Lukas and Lindstro, 1991; Chu et al., 2002). The excessive precipitation, river runoff and redistribution of the lowsalinity water by advection are in favor of the presence of BL. In the SEAS, the existence of the BL has been re-

2 246 GUO / J. Ocean Univ. China (Oceanic and Coastal Sea Research) : ported by several earlier studies. The winter onsoon current brings the low-salinity and low-teperature water fro the Bay of Bengal into the SEAS (Shetye et al., 1991; Rao and Sivakuar, 2003; Shara et al., 2010), which is considered to contribute to the foration of a BL with shallow salinity and density stratification and an inversion layer (Thadathil and Gosh, 1992; Shenoi et al., 1999). The heaviest precipitation on the eastern side of the Bay of Bengal and the Arabian Sea (Xie et al., 2006) is also favorable for the foration of BL. The BL thickness (BLT) is an iportant factor influencing the SST anoaly, as the thicker BL helps keeping the heat in the shallow ixed layer, leading to an increase in SST (Lewis et al., 1990; Vialard and Delecluse, 1998a, b). In addition, the existence of a thick BL prevents the ixed layer fro reaching the therocline, suppresses the sub-surface cold water fro getting into the surface layer and thus inhibits the surface cooling (Vialard and Delecluse, 1998a; Han et al., 2001; Annaalai et al., 2003; Qiu et al., 2012). The foration and developent of a BL in the SEAS also play an iportant role for SST variation in Arabian Sea (Shenoi et al., 2004; Durand et al., 2004). Masson et al. (2005) showed that the existence of BL and the subsurface inversion in the SEAS substantially contribute to the SST increase and the early onset of the south-west onsoon. Using an ocean general circulation odel (OGCM), De Boyer Montégut et al. (2007) found that heat accuulated in the barrier layer in the eastern Arabian Sea can war the surface layer by 0.4. In this study, it is proposed that the war SST is iportant to aintain the southeasterly wind anoaly in the tropical Indian Ocean during the foration of a positive IOD, and there is an interannual variation of BL in the SEAS, which contributes to the anoalous war SST in this region. Using Argo and SODA datasets the role of BL in the positive IOD is discussed in this paper as organized as follows. Section 2 introduces the data used in the study. Section 3 proposes the role of the BL during the developent of the IOD event and discusses the war SST and precipitation anoaly in the SEAS. The suary is in Section 4. 2 Data In the present study, the onthly ean wind at 10 and surface heat flux fro the NCEP/NCAR reanalysis products are adopted. This dataset has a spatial resolution of and covers the period fro 1950 to The onthly ean global precipitation dataset, Precipitation Reconstruction (Chen et al., 2003), is used to exaine the rainfall changes in the ocean, which is constructed on a horizontal grid over the globe for the period fro 1948 to the present. The upper-level ocean teperature and salinity are needed to study the developental process of the positive IOD events. Aong any available products, the onthly ean data fro the siple ocean data assiilation (SODA, v2.2.4) are selected for this study. The hori- zontal resolution of the dataset is and the vertical resolution is 10 in the upper-level (a total of 40 vertical layers, Carton and Giese, 2008). The wind data used to force this version of SODA are fro the 20th Century Atospheric Reanalysis products (Copo et al., 2011), which copare well with the NCEP/NCAR reanalysis, which is also used here, for both are based on surface observations and the NCEP Global Forecast Syste. The adequacy of using SODA data to study the Indian Ocean coupled dynaics is discussed in Xie et al. (2002). In addition, the teperature and salinity profiles fro Argo floats are used to investigate the IOD event in The Argo data have a 1 1 horizontal resolution and were collected fro 2005 to Because the IOD doinates the SST interannual variability, the decadal/interdecadal coponents are reoved through band-pass filters and only the signals of interannual variability (13 84 onths) are retained in order to extract the useful inforation related to IOD. 3 The Role of the Barrier Layer The SST anoaly presents a positive IOD event as an east (cold) to west (war) dipole ode. In Saji et al. (1999), the war SST anoaly exists in the SEAS during the developent of the positive IOD events. In this study the 7-year Argo and 49-year SODA data are used to illustrate the ipact of the BL anoaly on the anoalous war SST in the SEAS. The war SST in the SEAS and cold SST to the southeast of the SEAS induce the southeasterly wind anoaly. This process plays an iportant role in the developent of the IOD events. 3.1 Persistence of SST Anoaly in the SEAS Based on the IOD index defined by Saji et al. (1999), the standard deviation of SST difference between the western and eastern equatorial Indian Ocean should be larger than 1.5 as IOD coposite cases are identified. Based on this criteria, seven strong positive IOD events occurred in 1961, 1967, 1972, 1982, 1994, 1997 and 2006, respectively (Fig.1). Copared with the IOD cases in Saji et al. s study (1999), the 2006 IOD event is the first strong IOD event since Argo floats were fully deployed in the Indian Ocean. Fig.2 shows the evolution of SST and sea surface wind anoalies in 2006, together with the coposite results associated with the seven IOD events indicated in Fig.1. Only statistically significant results with a confidence level exceeding 90% are shown in Figs.2d, 2e and 2f. For all seven IOD years, an anoalous war SST occurs in the SEAS during June July, and stays there for the next several onths, accopanied with southeasterly winds. A noticeable difference is that war SST in the 2006 IOD year is also observed in the central southern Indian Ocean (Fig.2). Chowdary et al. (2009) found that a thick barrier layer propagating with the pronounced westward Rossby waves potentially contributed to this waring in the SEAS. According to the coposite results of SST, the SEAS is defined as the crucial region

3 GUO / J. Ocean Univ. China (Oceanic and Coastal Sea Research) : in this study, ost of which is within the western pole of IOD defined by Saji et al. (1999). The SST anoalies and the northwest-to-southeast dipole pattern, fored by the anoalous waring in the SEAS and cooling in the southeastern tropical Indian Ocean, are siilar when coparing the results with the Argo data (left colun) and the SODA data (right colun) in Fig.2. The SST patterns fro SODA are also siilar to those fro the Met Office Hadley Centre s sea ice and sea surface teperature dataset (HadISST1). To ensure the credibility of the Argo and SODA data used here, the SST patterns in the four strong IOD events (1982, 1994, 1997, and 2006) are further verified by using the daily data of the optiu interpolation sea surface teperature version 2 (OISST.V2) fro 1981 to the present (Reynolds et al., 2002). Such a SST pattern during the early developent of IOD results in strong southeasterly wind anoaly and the corresponding upwelling in the southeastern tropical Indian Ocean. As the anoalous war SST extends westward and southward starting fro August the anoalous Fig.1 Tie evolution of the noralized Indian Ocean Dipole ode index (DMI in ). The star arks the IOD year. Fig.2 The 2006 SST anoaly (shaded in ) and sea surface wind anoaly (arrows in s -1 ) fields fro (a) June July, (b) August Septeber and (c) October Noveber based on the Argo data. The sae are shown in (d) to (f) for the coposites of all seven positive IOD events based on SODA. The statistical significance of the analyzed anoalies is estiated by t-test. Anoalies of SST and surface wind exceeding 90% significance level are shown by the color shading and arrow, respectively. The black box indicates the crucial region in SEAS (0 10 N, 60 E 75 E).

4 248 GUO / J. Ocean Univ. China (Oceanic and Coastal Sea Research) : southeasterly winds are further strengthened. The existence of war SST anoaly in the SEAS plays a crucial role in enhancing the anoalous southeasterlies in the southeastern tropical Indian Ocean and the equatorial region. Next the echaniss to aintain the waring in the SEAS will be investigated. 3.2 The Role of the Barrier Layer in the SEAS In order to understand how the BL aintains the war SST anoaly in the SEAS, the BLT is calculated as the difference between the isotheral layer depth and the MLD, where the isotheral layer depth is defined as the depth at which teperature is 0.5 lower than SST (Monterey and Levitus, 1997), and the MLD is the depth where water density is kg -3 higher than that at the sea surface (Levitus, 1982; Huang and Qiu, 1994). A vertical linear interpolation is adopted due to the insufficient vertical resolution of SODA data, where either the MLD or isotheral layer depth lies between two standard levels. Using Argo floats data fro 2005 to 2010 and SODA data fro 1960 to 2008, the cliatology ean of BLT and the standard deviation in the tropical Indian Ocean duirng June-Noveber are calculated and shown in Fig.3. Fro both datasets, a thicker BL exists in the southeastern tropical Indian Ocean, the Bay of Bengal, and the Arabian Sea, although the Argo data shows a ean BLT of 10 and the SODA data shows a BLT of 5 in the SEAS. The variation of BLT in the SEAS is significant and is alost equal to its cliatology ean. The teperature and salinity fro SODA copares well with those fro the Argo data. Fig.3 Barrier layer thickness (contours in ) and its standard deviation (color shading in ) in June Noveber based on (a) Argo data averaged fro 2005 to 2010 and (b) SODA averaged fro 1960 to The crucial region in SEAS (0 10 N, 60 E 75 E) is outlined by dashed line. Previous studies (Annaalai et al., 2003; Masson et al., 2005) have exained the waring effect of BL on SST in the Indian Ocean. Here it is proposed that the variation of the BLT ay influence the SST anoaly in the SEAS. In order to confir the hypothesis, the ixed layer teperature equation is considered (e.g., Qiu, 2000; Liu et al., 2005): T t Q c h V T w ( T T ) h net p e d. (1) Using the perturbation ethod, Eq. (1) can be written as: T ' Q w ( T Td) ( ) h ' t c h h c net e 2 2 p Q ' p h ( V ' T V T' ) w'( e T Td) we ( T' T') d ( ). (2) h h Aong the four ters on the right hand side of Eq. (2), the first ter is the contribution of MLD anoaly to the SST variation anoaly; the second, third, and fourth ters are the contributions of the net air sea heat flux anoaly, the current anoaly and horizontal gradient of SST anoaly, and the vertical velocity anoaly and vertical teperature gradient anoaly, respectively. ρ and c p in the equation are the sea water density and specific heat at constant pressure, respectively; Q net is the ean net air sea heat flux; V is the ean current velocity vertically averaged over the ixed layer, which is coposed of surface geostrophic current and Ekan flow; w e is the ean entrainent velocity; h is the ean ixed layer depth; T is the ean teperature in the ixed layer; T is the ean teperature beneath the ixed d ' net layer; Q, V ', h ', T ', w ' e, T ' d are the respective anoalies. The four ters averaged over the SEAS for each onth in 2006 are shown in Fig.4a. Aong those ters, the first ter is the largest, except in June, and reaches its axiu (0.33 on -1 ) in October (Figs.4a, and 4b), and the fourth ter is the sallest and can be neglected. The large values of the first ter indicate that the positive SST anoaly ainly coes fro the contribution of negative MLD anoaly, for a thinner ixed layer has a lower heat capacity. In general, the contribution of negative MLD anoaly (the first ter) is doinant in the SST waring during the developent period of the positive IOD in Corresponding to the negative MLD anoalies, there are positive BLT anoalies (Fig.4b). The coposite of each of the four ters during the seven positive IOD events is shown in Fig.4c. The above discussion on the 2006 Argo data applies to these positive IOD events as well: the positive SST anoaly in the SEAS is ainly due to the contribution of negative MLD anoaly in Fig.4d, where it shows the corresponding positive BLT anoaly, and the absolute values of the BLT are alost the sae as the absolute MLD anoaly in the coposite results. Because the MLD decreases by about

5 GUO / J. Ocean Univ. China (Oceanic and Coastal Sea Research) : % 20% fro its cliatology, the effect of net air sea heat flux on the SST is aplified, for a thinner ML has a lower heat capacity. The net heat fluxe reanalysis fro NCEP copares well with the objectively analyzed air sea fluxes (OAFlux, Praveen-Kuar et al., 2012) and OAFlux (Yu et al., 2008). Corresponding to both for the 2006 IOD event and the coposite of seven IOD events, the MLD shows negative anoalies, while the BLT positive anoalies. A shallower ML is the ain echanis for the persistence of positive SST anoalies in the SEAS during the IOD foration period. Morioka et al. (2010, 2011) pointed out the iportance of the MLD in SST anoaly and the findings here are consistent with their results. Fig.4 (a) The teperature variation anoaly ( on -1 ), including the MLD anoaly (1st ter, dark bar), the sea surface net heat flux anoaly (2nd ter, dark gray bar), the current anoaly and the horizontal gradient of SST anoaly (3rd ter, light gray bar), and the vertical velocity anoaly and vertical teperature gradient anoaly (4th ter, white bar); (b) The ixed layer depth anoaly (MLDA in, dark gray bar) and the barrier layer thickness anoaly (BLTA in, white bar) averaged in the SEAS (0 10 N, 60 E 75 E) fro June to Noveber. (a) and (b) are based on the 2006 Argo data; (c) and (d) show the sae variables as (a) and (b), respectively, for the coposite of seven positive IOD events fro SODA. 3.3 The Ocean-Atosphere Interaction Processes Aong Wind, Precipitation, BL, and SST in the SEAS In the tropics, the BL usually appears as a low-salinity water layer between isotheral layer and ML after heavy precipitation or low-salinity advection fro other areas. The results of the NCEP/NCAR reanalysis precipitation indicate that the positive precipitation anoaly begins to appear in the SEAS in August and becoes larger gradually afterward (Figs.5b to 5c). A coposite of precipitation anoaly appearing in the seven IOD years (Figs.5d to 5f) shows that in the positive IOD events the precipitation has a positive anoaly in the SEAS in June July and this anoaly increases in agnitude with tie, which agrees well with changes in SST (Fig.2) and BLT (Fig.4d). Opposite to the heavy precipitation in the SEAS, the precipitation reduces greatly in the eastern Indian Ocean fro June to Noveber. The precipitation pattern of the four strong IOD events (1982, 1994, 1997 and 2006) agrees well with the Global Precipitation Cliatology Project version 2 (GPCP.V2.2) cobined precipitation data set (Adler et al., 2003). The positive precipitation anoaly is iportant for a positive anoaly of the BLT in the SEAS. Other studies also show that fresh water coing fro the Bay of Bengal is iportant for salinity budget in the SEAS (Jensen, 1991; Shetye et al., 1991; Shenoi et al., 1999; Rao and Sivakuar, 1999, 2003). In order to exaine the relationships aong precipitation, SST, BL and MLD, the noralized tie series of Septeber-Noveber precipitation anoaly (PreA), SST anoaly (SSTA), BLT anoaly (BLTA) and MLD anoaly (MLDA) are averaged in the SEAS fro 1960 to 2008 and are shown in Fig.6. The PreA is strongly correlated with BLTA during the peak IOD period (Septeber Noveber), with a positive correlation coefficient of 0.75 and a significance level exceeding 95%. This strong correlation further indicates that ore precipitation will increase the thickness of BL. Corresponding to the thicker BL, the shoaling of ML will war the SST based on the early discussion. It is also found that war SST could in turn increase the BLT. The relationship between SSTA and BLA (MLDA) is exained and the correlation coefficient between the is 0.47 ( 0.48) with a significance level exceeding 95%. War SST in the SEAS and cold SST in the southeastern tropical Indian Ocean for an SST gradient, which provides a favorable condition for

6 250 GUO / J. Ocean Univ. China (Oceanic and Coastal Sea Research) : the occurrence of the anoalous southeasterly winds (Fig.2). The winds can bring abundant oisture fro the southeastern Indian Ocean, and enhance the precipitation. The correlation coefficient between SSTA and PreA is 0.46 with a confidence level of 95%. The excessive precipitation results in the BL thickening and the ML shoaling. The correlation coefficients between the four variables and the IOD index of SON are 0.79, 0.72, 0.68 and 0.47 with a significance level exceeding the 95%, respectively, indicating a close relationship between those variables and the IOD event. Saji et al. (1999) pointed out the great changes of zonal wind over the equatorial central and eastern Indian Ocean during the IOD events. Our results further illustrate that there is an ocean-atosphere interaction process in the SEAS aong SST, wind, precipitation, BL, and ML during a positive IOD event. Along with other echaniss the BL plays an iportant role in the developent of IOD. Fig.5 The precipitation anoaly fields (contours in d -1 ) in (a) June July, (b) August Septeber and (c) October Noveber using the 2006 Argo data. The sae fields are shown in (d) (f) for the coposite of seven positive IOD events fro SODA and the shaded area has a significance level exceeding 90%. The box indicates the crucial region in SEAS (0 10 N, 60 E 75 E). Fig.6 The noralized tie series of the precipitation anoaly (PreA, black solid line), sea surface teperature anoaly (SSTA, gray solid line), barrier layer thickness anoaly (BLTA, dashed line), and ixed layer depth anoaly (MLDA, gray bar) averaged in the SEAS in Septeber Noveber.

7 GUO / J. Ocean Univ. China (Oceanic and Coastal Sea Research) : Suary and Discussion Using data fro the Argo floats and SODA reanalysis, the coposite analysis is conducted to investigate the developent processes of positive IOD events. It is found that warer SST appears in June July in the SEAS (0 10 N, 60 E 75 E), ost of which is in the western pole of the IOD defined by Saji et al. (1999). This anoalous war SST extends westward and southward, and can persist for several onths. The positive teperature anoaly in the SEAS is ainly due to the MLD decrease (the first ter in the ixed layer teperature equation), especially fro August to Noveber. Because the MLD is about 15% 20% thinner coparing with its cliatology, the effect of net air sea heat flux on the SST is aplified, for a thinner ML has a lower heat capacity. The contribution of net heat flux (the second ter) and the total contribution of the current anoaly and horizontal gradient of SST anoaly (the third ter) are positive in soe onths but are saller than the contribution of the MLD anoaly (the first ter). The total contribution of vertical velocity anoaly and vertical teperature gradient anoaly (the fourth ter) is so sall that their effect is negligible. It is shown that ML and BL are iportant in adjusting ocean-atosphere interactions and thus affecting weather and cliate. Waring in the SEAS caused by the thickening of BLT (shoaling of MLD) enhances the precipitation and increases the southeast-northwest SST gradient with the cold SST in the southeastern Indian Ocean. The large SST gradient intensifies the anoalous southeasterly winds, which brings ore oisture to the SEAS. The excessive precipitation reduces the salinity, increases the stratification near the surface, and leads to a decrease in MLD and an increase in BLT, which then contributes to a warer SST in the SEAS. This ocean-atosphere interaction process aong wind, precipitation, BL and SST plays an iportant role during the developent of positive IOD events in the SEAS. Acknowledgeents Coents by the two anonyous reviewers greatly helped iprove the anuscript. Thanks to Profs. S.-P. Xie and W. Han for their coents on the anuscript. Argo data were collected and ade available by the International Argo Progra of the Global Ocean Observing Syste ( org). This study is supported by the National Basic Research Progra of China (2012CB955602), Ministry of Science and Technology of China (National Key Progra for Developing Basic Science 2010CB428904), the NSFC ( , , ), and the 111 Project of China (Progra of Introducing Talents of Discipline to Universities No. B07036). References Adler, R. F., Huffan, G. J., Chang, A., Ferraro, R., Xie, P., Janowiak, J., Rudolf, B., Schneider, U., Curtis, S., Bolvin, D., Gruber, A., Susskind, J., and Arkin, P., The Version 2 global precipitation cliatology project (GPCP) onthly precipitation analysis (1979 present). Journal of Hydroeteorology, 4: Annaalai, H., Murtugudde, R., Potera, J., Xie, S., Liu, P., and Wang, B., Coupled dynaics over the Indian Ocean: Spring initiation of the zonal ode. Deep-Sea Research II, 50: Bjerknes, J., Atospheric teleconnections fro the equatorial Pacific. Monthly Weather Review, 97: Carton, J. A., and Giese, B. S., A reanalysis of ocean cliate using siple ocean data assiilation (SODA). Monthly Weather Review, 136: Chen, M., Xie, P., Janowiak, J. E., Arkin, P. A., and Sith, T. M., Reconstruction of the oceanic precipitation fro 1948 to the present. The AMS 14th Syposiu on Global Changes and Cliate Variations. Aerican Meteorological Society, Long Beach, CA. Chowdary, J. S., Gnanaseelan, C., and Xie, S.-P., Westward propagation of barrier layer foration in the Rossby wave event over the tropical southwest Indian Ocean. Geophysical Research Letters, 36, L04607, DOI: / 2008GL Chu, P. C., Liu, Q., Jia, Y., and Fan, C., Evidence of a Barrier Layer in the Sulu and Celebes Seas. Journal of Physical Oceanography, 32 (11): Copo, G. P., Whitaker, J. S., Sardeshukh, P. D., Matsui, N., Allan, R. J., Yin, X., Gleaso B. E., Vose, R. S., Rutledge, G., Besseoulin, P., Bronniann, S., Brunet, M., Crouthael, R. I., Grant, A. N., Groisan, P. Y., Jones, P. D., Kruk, M. C., Kruger, A. C., Marshall, G. J., Maugeri, M., Mork, H. Y., Nordli, Ø., Ross, T. F., Trigo, R. M., Wang, X. L., Woodruff, S. D., and Worley, S. J., The twentieth century reanalysis project. Quarterly Journal of the Royal Meteorological Society, 137: 1-28, DOI: /qj.776. De Boyer Montégut, C., Vialard, J., Shenoi, S. S. C., Shankar, D., Durand, F., Ethé, C., and Madec, G., Siulated seasonal and interannual variability of ixed layer heat budget in the northern Indian Ocean. Journal of Cliate, 20: , DOI: / JCLI Durand, F., Shetye, S. R., Vialard, J., Shankar, D., Shenoi, S. S. C., Ethé, C., and Madec, G., Ipact of teperature inversions on SST evolution in the south-eastern Arabian Sea during the pre-suer onsoon season. Geophysical Research Letters, 31, L01305, DOI: /2003GL Godfrey, J. S., and Lindstro, E. J., The heat budget of the equatorial western Pacific surface ixed layer. Journal of Geophysical Research, 94: Halkides, D. J., Han, W., and Webster, P. J., Effects of the seasonal cycle on the developent and terination of the Indian Ocean Zonal Dipole Mode. Journal of Geophysical Research, 111, C12017, DOI: /2005JC Han,W., McCreary, J. P., and Kohler, K. E., Influence of precipitation-evaporation and Bay of Bengal rivers on dynaics, therodynaics, and ixed layer physics in the upper Indian Ocean. Journal of Geophysical Research, 106: Huang, B., and Kinter III, J. L., Interannual variability in the tropical Indian Ocean. Journal of Geophysical Research, 107, 3199, DOI: /2001JC Huang, R. X., and Qiu, B., Three-diensional structure of the wind-driven circulation in the subtropical North Pacific. Journal of Physical Oceanography, 24: Jensen, T. G., Modeling the seasonal undercurrents in the

8 252 GUO / J. Ocean Univ. China (Oceanic and Coastal Sea Research) : Soali Current syste. Journal of Geophysical Research, 96 (C12): , DOI: /91JC Le Blanc, J. L., and Boulanger, J. P., Propagation and reflection of long equatorial waves in the Indian Ocean fro TOPEX/POSEIDON data during the period. Cliate Dynaics, 17: Levitus, S., Cliatological Atlas of the World Ocean. National Oceanic and Atospheric Adinistration. NOAA Professional Paper-13, 173pp. Lewis, M. R., Carr, M., Feldan, G., Esaias, W., and Mc-Clain, C., Influence of penetrating solar radiation on the heat budget of the equatorial Pacific Ocean. Nature, 347: Li, T., Wang, B., Chang, C. P., and Zhang, Y., A theory for the Indian Ocean Dipole zonal ode. Journal of the Atospheric Sciences, 60: Liu, Q., Xie, S. P., Li, L., and Maxienko, N. A., Ocean theral advective effect on the annual range of sea surface teperature. Geophysical Research Letters, 32, L24604, DOI: /2005GL Lukas, T., and Lindstro, E., The ixed layer of the western equatorial Pacific Ocean. Jouranl of Geophysical Research, 96: Luo, J. J., Zhang, R. C., Behera, S. K., Masuoto, Y., Jin, F. F., Lukas, R., and Yaagata, T., Interaction between El Niño and Extree Indian Ocean Dipole. Journal of Cliate, 23: , DOI: /2009JCLI Masson, S, Luo, J. J., Madec, G., Vialard, J., Durand, F., Gualdi, S., Guilyardi, E., Behera, S., Delecluse, P., Navarra, A., and Yaagata, P., Ipact of barrier layer on winter-spring variability of the southeastern Arabian Sea. Geophysical Research Letter, 32, L07703, DOI: /2004GL Monterey, G., and Levitus, S., Seasonal variability of ixed layer depth for the world ocean. NOAA Atlas NESDIS 14. U.S. Gov. Printing Office, Washington D.C., 96pp. Morioka, Y., Tozuka, T., and Yaagata, T., Cliate variability in the Southern Indian Ocean as revealed by self- organizing aps. Cliate Dynaics, 35: , DOI: /s x. Morioka, Y., Tozuka, T., and Yaagata, T., On the growth and decay of the subtropical dipole ode in the South Atlantic. Journal of Cliate, 24: , DOI: / 2011JCL Murtugudde, R., McCreary, J. P., and Busalacchi, A. J., Oceanic processes associated with anoalous events in the Indian Ocean with relevance to Journal of Geophysical Research, 105: Praveen Kuar, B., Vialard, J., Lengaigne, M., Murty, V. S. N., and McPhaden, M. J., TropFlux: air sea fluxes for the global tropical oceans-description and evaluation. Cliate Dynaics, 38: , DOI: /s Qiu, B., Interannual variability of the Kuroshio extension syste and its ipact on the wintertie SST field. Journal of Physical Oceanography, 30: Qiu, Y., Cai, W., Li, L., and Guo, X., Argo profiles variability of barrier layer in the tropical Indian Ocean and its relationship with the Indian Ocean Dipole. Geophysical Research Letters, 39, L08605, DOI: /2012GL Rao, R. R., and Sivakuar, R., On the possible echaniss of the evolution of a ini-war pool during the presuer onsoon season and the onset vortex in the southeastern Arabian Sea. Quarterly Journal of the Royal Meteorological Society, 125: Rao, R. R., and Sivakuar, R., Seasonal variability of sea surface salinity and salt budget of the ixed layer of the north Indian Ocean. Journal of Geophysical Research, 108 (C1), 3009, DOI: /2001JC Reynolds, R. W., Rayner, N. A., Sith, T. M., Stokes, D. C., and Wang, W., An iproved in situ and satellite SST analysis for cliate. Journal of Cliate, 15: Saji, N. H., Goswai, B. N., Vinayachandran, P. N., and Yaagata, T., A dipole ode in the Tropical Indian Ocean. Nature, 401: Shara, R., Agarwal, N., Moin, I. M., Basu, S., and Agarwal, V. K., Siulated sea surface salinity variability in the Tropical Indian Ocean. Journal of Cliate, 23: , DOI: JCLI Shenoi, S. S. C., Shankar, D., and Shetye, S. R., On the sea surface teperature high in the Lakshadweep Sea before the onset of the southwest onsoon. Journal of Geophysical Research, 104 (C7): Shenoi, S. S. C., Shankar, D., and Shetye, S. R., Reote forcing annihilates barrier layer in southeastern Arabian Sea. Geophysical Research Letters, 31, L05307, DOI: / 2003GL Shetye, S. R., Gouveia, A. D., Shenoi, S. S. C., Michael, G. S., Sundar, D., Aleida, A. M., and Santana, K., The coastal current off western India during the northeast onsoon. Deep-Sea Research (Part A), 38: Thadathil, P., and Gosh, A. K., Surface layer teperature inversion in the Arabian Sea during winter. Journal of Oceanography, 48: Vialard, J., and Delecluse, P., 1998a. An OGCM study for the TOGA decade. Part I. Role of salinity in the physics of the western Pacific fresh pool. Journal of Physical Oceanography, 28: Vialard, J., and Delecluse, P., 1998b. An OGCM study for the TOGA decade. Part II. Barrier-layer foration and variability. Journal of Physical Oceanography, 28: Wang, B., Wu, R., and Li, T., Atosphere-war ocean interaction and its ipact on Asian-Australian onsoon variation. Journal of Cliate, 16: Webster, P. J., Moore, A. M., Loschnigg, J. P., and Leben, R. R., Coupled ocean atosphere dynaics in the Indian Ocean during Nature, 401: Xie, S. P., Annaalai, H., Schott, F. A., and McCreary, J. P., Structure and echaniss of South Indian Ocean cliate variability. Journal of Cliate, 15: Xie, S. P., Xu, H., Saji, N. H., Wang, Y., and Liu, W. T., Role of narrow ountains in large-scale organization of Asian onsoon convection. Journal of Cliate, 19: Yaagata, T., Behera, S. K., Rao, S. A., Guan, Z., Ashok, K., and Saji, H. N., The Indian Ocean Dipole: a physical entity. CLIVAR Exchanges, 24: Yaagata, T., Behera, S. K., Rao, S. A., Guan, Z., Ashok, K., and Saji, H. N., Coents on Dipoles, teperature gradient, and tropical cliate anoalies. Bulletin of the Aerican Meteorological Society, 84: Yu, L., Jin, X., and Weller, R. A., Multidecade global flux datasets fro the objectively analyzed air sea fluxes (OAFlux) project: Latent and sensible heat fluxes, ocean evaporation, and related surface eteorological variables. Woods Hole Oceanographic Institution OAFlux Project Technical Report (OA ), Woods Hole. Massachusetts. 64pp. Zheng, X. T., Xie, S. P., Vecchi, G. A., Liu, Q., and Hafner, J., Indian Ocean Dipole response to global waring: analysis of ocean atospheric feedbacks in a coupled odel. Journal of Cliate, 23: (Edited by Xie Jun)