A Refined Perspective on Indian Ocean Dipole–Driven Surface Salinity Changes and the Salinity Dipole Index

A Refined Perspective on Indian Ocean Dipole–Driven Surface Salinity Changes and the Salinity Dipole Index | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article A Refined Perspective on Indian Ocean Dipole–Driven Surface Salinity Changes and the Salinity Dipole Index Kokulathasan Thilaksanth, Gayan Pathirana, Kyung-Min Noh, Aoyun Xue, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8401709/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 4 You are reading this latest preprint version Abstract The Indian Ocean Dipole (IOD) is a dominant coupled ocean–atmosphere mode in the tropical Indian Ocean, characterized by opposing sea surface temperature anomalies (SSTAs) between the western and southeastern regions. In addition to its well-established climatic and biological impacts, the IOD strongly modulates sea surface salinity (SSS), a key variable governing ocean circulation and air–sea interactions. However, the nature of IOD-induced salinity changes and the appropriate formulation of a salinity dipole index (SDI) remain debated, particularly across different IOD phases. Previous studies have proposed contrasting definitions of salinity dipole regions: one, based on high-resolution eddy-resolving model simulations, identifies the dipole between the Central Equatorial Indian Ocean (CEIO: 70°E–90°E, 5°S–5°N) and the Sumatra–Java Coast (SJC: 100°E–110°E, 13°S–3°S); while another, derived from satellite-based SSS observations, aligns with the conventional IOD regions. Here, we revisit the IOD–SSS relationship and reassess the SDI using long-term observational data and historical simulations from a Coupled Model Intercomparison Project Phase 6 (CMIP6) model datasets. Our results show that the IOD-induced salinity dipole emerges most clearly during boreal autumn but is not co-located with the conventional IOD regions. Instead, it consistently develops between the CEIO and SJC, particularly during strong positive IOD events, exhibiting low salinity in the CEIO and high salinity along the SJC. This dipole structure is primarily controlled by IOD-driven surface water advection and freshwater flux anomalies. Based on these findings, we propose a refined definition of the salinity dipole regions and SDI, which provides a more robust and consistent representation of salinity variability during IOD events. Sea Surface Salinity Indian Ocean Dipole salinity dipole index Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Figure 7 Figure 8 1. Introduction Ocean salinity is a fundamental physical property of the ocean that plays a crucial role in large-scale circulation, air–sea interactions, and climate variability (Stammer et al., 2021 ). Salinity influences thermohaline circulation (Hu et al., 2019 ) and ocean stratification (Balaguru et al., 2016 ), acting as a key regulator of global climate systems. Variations in salinity further impact marine ecosystems by altering ocean productivity, food web dynamics, and marine species distribution (Doney et al., 2012 ). At a regional scale, surface salinity has a strong influence on the thickness of barrier layers (Bosc et al., 2009 ) and the depth of ocean mixed layers (de Boyer Montégut et al., 2004 ), as well as on subduction processes, thereby linking ocean thermodynamics with the hydrological cycle (Lagerloef, 2002 ). Further, sea surface salinity (SSS) reflects the combined influence of freshwater fluxes (evaporation, precipitation, and river discharge), ocean advection, entrainment, and vertical mixing (Delcroix et al., 1996 ; Perigaud et al., 2003 ; Sandeep et al., 2018 ). SSS therefore represents both a driver of ocean dynamics and a sensitive indicator of freshwater fluxes, ocean circulation, and climate variability. The Indian Ocean (IO) is one of the most hydrographically diverse basins, with a SSS distribution that reflects strong regional contrasts. In particular, the Arabian Sea (AS) in the northern IO is characterized by high salinity due to net evaporation (Kumar & Prasad, 1999 ), while the Bay of Bengal (BoB) remains persistently fresh due to river discharge and high precipitation (Rao & Kumar, 1991 ). On a seasonal timescale, the monsoon seasons have a significant influence on the spatial variability of surface winds and currents in the northern IO (Jinadasa et al., 2020 ; Schott & McCreary, 2001 ). In response to the shift in winds and surface currents from the southwest monsoon (SWM) to the northeast monsoon (NEM), water is exchanged between the AS and BoB basins surrounding Sri Lanka. This reflects a mixture of low and high salinity waters on a seasonal timescale. In contrast, the Central Equatorial Indian Ocean (CEIO) exhibits a semi-annual salinity cycle linked to the monsoon system (Yuhong et al., 2013 ). In addition to these seasonal and regional processes, the interannual and decadal variability of SSS in the IO is significantly influenced by coupled climate modes, particularly the Indian Ocean Dipole (IOD) and the El Niño–Southern Oscillation (ENSO) (Du & Zhang, 2015 ; Jensen, 2001 ; Rao & Sivakumar, 2003 ; Saji & Yamagata, 2003 ). Among the tropical air-sea coupled climate modes, the IOD is the dominant interannual mode of climate variability in the tropical IO. It is defined by opposing sea surface temperature anomalies (SSTAs) between the western (10° S–10° N, 50° E–70° E) and southeastern equatorial (10° S–0°, 90° E–110° E) regions (Saji et al., 1999 ). Developing during the boreal summer and peaking in the autumn, the IOD influences rainfall, agriculture, fisheries, and even human health across the IO-rim countries and beyond (Ashok et al., 2001 ; Behera et al., 2005 ; Yuan & Yamagata, 2015 ). Notably, IOD events are associated not only with SST anomalies, but also with pronounced salinity anomalies (SSSA), which in turn feed back into the coupled system, modulating IOD evolution (Rao & Sivakumar, 2003 ). Several studies have highlighted robust salinity signatures during IOD phases. Positive IOD (pIOD) events are characterized by freshening in the CEIO and salinification along the Sumatra–Java Coast (SJC), whereas negative IOD (nIOD) events exhibit opposite patterns (Li et al., 2016 ; Subrahmanyam et al., 2011 ; Thompson et al., 2006 ). These contrasting anomalies form a distinct salinity dipole. To quantify this phenomenon,Li et al. ( 2016 ) introduced the Dipole Mode Index of Salinity (DMIS), which is defined as the difference in SSSA between the CEIO and the SJC. More recently,Shi & Wang, ( 2025 ) introduced a novel salinity dipole mode index (SDMI) based on the conventional SST-defined IOD regions. However, this latter approach assumes that salinity variability mirrors SST anomalies spatially, a premise that has been challenged by observational and modelling studies alike. Such inconsistencies in defining the salinity dipole index are like debates surrounding other dipole indices. For instance, Shi & Wang ( 2021 , 2022 ) introduced a biological IOD (BIOD) and a related index based on conventional IOD regions. However, subsequent research demonstrated that biological responses were more accurately represented by alternative regions, resulting in the development of a refined biological dipole index (BDI) (Pathirana et al., 2024 a, 2024 b; Abeywickrama et al., 2025 ). This suggests that the accuracy of dipole indices in representing the underlying variability is critically determined by the choice of index regions. In the case of SSS, it remains an open question whether the dipole is best captured by the CEIO–SJC framework or by conventional IOD regions. Against this backdrop, revisiting the definition of a salinity dipole index is essential. A robust, physically consistent Salinity Dipole Index (SDI) would advance our understanding of salinity variability during IOD events and improve the representation of freshwater fluxes, surface circulation, and air–sea coupling in the tropical IO. Furthermore, refining the SDI has implications for climate prediction, given that SSS is increasingly recognized as a key precursor and modulator of ocean–atmosphere feedback (Chen et al., 2019 ). In this study, we re-evaluate the relationship between the IOD and SSS using long-term observational and historical simulations from a Coupled Model Intercomparison Project Phase 6 (CMIP6) model datasets, proposing a refined formulation of the salinity dipole regions and index. This refined SDI provides a clearer and more consistent measure of salinity variability, offering a critical tool for both fundamental climate research and applied prediction efforts. 2. Data and Methods To investigate the influence of the IOD on SSS and to refine the salinity dipole index (SDI) definition, we analyzed oceanic and atmospheric datasets from 1993 to 2023. We obtained monthly SST, SSS, surface current, and subsurface temperature and salinity from the Copernicus Marine Environment Monitoring Service (CMEMS) reanalysis products (CMEMS, 2024). Atmospheric variables, including 10 m winds, precipitation, and evaporation, were obtained from the ERA5 reanalysis, produced by the European Centre for Medium-Range Weather Forecasts (ECMWF; Hersbach, 2023 ). In addition to observational data, historical simulations from the CMIP6 were analyzed using 13 CMIP6 models selected based on data availability (Table S2). Details of each dataset, including their spatial resolution, temporal coverage and access links, are summarized in Supplementary Table 3. To isolate variability, we calculated monthly anomalies by removing the mean seasonal cycle from each field. Our analysis proceeded in three stages. First, we examined large-scale salinity variability in relation to the dominant climate modes. We then applied multiple linear regression to regress SSS and SST anomalies separately onto the Dipole Mode Index (DMI; September–November, SON) and the Niño 3.4 index (December–February, DJF). This enabled us to determine the relative influence of IOD and ENSO on IO salinity variability. Based on these results, subsequent analyses focused on the SON season, when IOD impacts are strongest. For the SON season, we computed the SSS climatology and interannual standard deviation to characterize the baseline state and variability. Positive and negative IOD years were defined as those with normalized DMI values exceeding ± 1σ (Li et al., 2016 ). Composite SSS anomalies were then constructed to assess dipole patterns associated with pIOD and nIOD events. Second, we compared alternative definitions of the salinity dipole. Time series of SSS anomalies were extracted from the Central Equatorial Indian Ocean (CEIO: 70°E–90°E, 5°S–5°N) and the Sumatra–Java Coast (SJC: 100°E–110°E, 13°S–3°S) regions. These were then used to compute a CEIO–SJC salinity index. This was then contrasted with the Salinity Dipole Mode Index (SDMI) proposed by Shi and Wang ( 2025 ), which follows the conventional IOD regions. Correlation analyses were conducted between DMI, SDMI and CEIO–SJC–based SDI to evaluate which formulation more robustly captures salinity variability during IOD phases. Thirdly, we investigated the physical mechanisms driving the salinity dipole. Multiple regression analyses were performed on SON SSS anomalies in relation to key forcing fields, such as precipitation, evaporation, surface currents, and halocline depth. Halocline depth was estimated as the depth of the maximum vertical salinity gradient, computed from subsurface salinity profiles. To further quantify the relative roles of surface fluxes and ocean dynamics, we performed a mixed-layer salt budget analysis in accordance with the methodology of Li et al. ( 2016 ). The salt budget equation is expressed as follows (Sun et al., 2022 ): $$\:\frac{\partial\:\varvec{S}}{\partial\:\varvec{t}}=\:-\left(\left.\varvec{u}\frac{\:\partial\:\varvec{S}}{\partial\:\varvec{x}}+\varvec{v}\frac{\partial\:\varvec{S}}{\partial\:\varvec{y}}\right)\right.-\varvec{S}\frac{\left(\varvec{P}-\varvec{E}\right)}{\varvec{h}}+\:\varvec{\epsilon\:}$$ Where S is the mixed-layer salinity, t is time, u and v are the zonal and meridional velocity components, P and E are the precipitation and evaporation, and h is the mixed-layer depth, and ( \(\:\varvec{\epsilon\:}\) ) is the residual term representing unresolved processes. The mixed-layer depth was defined using the temperature criterion (de Boyer Montégut et al., 2004 ). A summary of index definitions, acronyms, and regional boundaries used in this study is summarized in Table 1 for reference. Table 1 Information on index regions, definitions, and acronyms used in the present study. IO Indian Ocean IOD Indian Ocean Dipole DMI Dipole Mode Index. DMI is defined as the difference in SST anomalies between the western Indian Ocean (10° S–10° N, 50° E–70° E) and the southeastern Indian Ocean (10° S–0°, 90° E–110° E, Saji et al., 1999 ). WIO Western Indian Ocean (10 0 S-10 0 N, 50 0 E-70 0 E) SEIO Southeastern Indian Ocean (10 0 S-0 0 , 90 0 E-110 0 E) SDMI Salinity Dipole Mode Index (Shi & Wang, 2025 ). SDMI is defined as the difference of average SSS anomaly in the Western Indian Ocean (10 0 S-10 0 N, 50 0 E-70 0 E) and the Southeastern Indian Ocean (10 0 S-0 0 , 90 0 E-110 0 E). CEIO Central Equatorial Indian Ocean (5 0 S–5 0 N, 70 0 E–90 0 E) SJC Sumatra Java Coast (13 0 S–3 0 N, 100 0 E–110 0 E) SDI* A salinity dipole index based on the SSS anomalies between the central equatorial Indian Ocean (5 0 S–5 0 N, 70 0 E–90 0 E) and the region off the Sumatra–Java coast (13 0 S–3 0 N, 100 0 E–110 0 E) was first proposed as the Dipole Mode Index for Salinity (DMIS) by (Li et al., 2016 ). In the present study, we introduce a refined Salinity Dipole Index (SDI), to more clearly quantify the SSS dipole pattern associated with the IOD. The SDI is defined as the difference between the average SSS anomaly over the CEIO and that over the SJC region. Compared to the previously defined SDMI, the SDI more directly captures the opposing salinity anomalies in the CEIO and SJC. 3. Results 3.1. Dipole pattern in Sea Surface Salinity During the IOD peak season, the mean state of SSS shows clear regional differences north of 20°S, reflecting the combined effects of air–sea fluxes, river discharge, and ocean circulation (see Fig. S1 a). The Bay of Bengal (BoB) and the eastern Indian Ocean (EIO) have relatively fresh surface waters, mainly due to more precipitation than evaporation and a lot of freshwaters coming from major rivers (Rao & Kumar, 1991 ; Li et al., 2021 ). In contrast, the Arabian Sea (AS) has high salinity due to persistent net evaporation (Kumar & Prasad, 1999 ). This large-scale salinity gradient highlights the complex interaction between freshwater fluxes, ocean mixing, and circulation processes. Notably, a high-salinity tongue extends eastward along the equator (Fig. S1 a), formed by the eastward advection of saline waters by the equatorial jet (Wyrtki, 1973 ). Meanwhile, a low-salinity tongue near 12°S (Fig. S1 a) originates from the westward advection of fresher waters by the South Equatorial Current. These current transport low-salinity waters from the EIO and the Indonesian Throughflow (Yuhong et al., 2013 ; Zhang et al., 2016 ). The interannual variability of SSS, represented by its standard deviation (Fig. S1 b), reveals distinct spatial patterns. The northern BoB exhibits strong variability, primarily due to variable river runoff, while regions along the east coast of India, the Andaman Sea, and the central equatorial Indian Ocean (CEIO) demonstrate pronounced fluctuations linked to large-scale climate phenomena such as the Indian Ocean Dipole (IOD) and the El Niño–Southern Oscillation (ENSO) (Chen et al., 2024 ). These patterns emphasize the dynamic nature of surface salinity and its sensitivity to local hydrological inputs and basin-scale climatic variability. Since IOD and ENSO are the major climate modes influencing SSS in tropical IO, we applied a multiple linear regression analysis to isolate their respective impacts. The regression patterns representing IOD-driven variations in SST, surface circulation, and SSS during the peak IOD phase in boreal autumn are presented in Fig. 1 , while the ENSO-related effects are shown separately in Fig. S2. In this paper, we focus primarily on the IOD-induced changes, as our regression framework effectively removes ENSO-induced changes to reveal the independent IOD signal. The IOD regression pattern clearly depicts the canonical east-west dipole structure, with significant cooling in the SEIO and warming in the WIO (Fig. 1 a). These temperature anomalies are accompanied by intensified southeasterly winds, driven by the SST-induced pressure gradient force. The resulting surface wind anomalies promote enhanced upwelling and mixing in the SEIO and downwelling in the WIO, further amplifying the SST contrast. The strengthened easterly winds transport cooler, drier air from the SEIO toward the warmer, convectively active WIO, reinforcing the thermodynamic and dynamic coupling between ocean and atmosphere. This feedback loop, characteristic of the positive IOD phase, exemplifies the development of the Bjerknes feedback mechanism that sustains and amplifies IOD events. To further assess how the IOD modulates SSS across tropical IO, we examined the regression patterns of SSS anomalies associated with IOD (Fig. 1 b). During pIOD events, SSS anomalies increase markedly in the SJC and eastern Indian coastal regions, while negative anomalies dominate the CEIO. The opposite pattern emerges during nIOD events, consistent with previous studies (Thompson et al., 2006 ; Subrahmanyam et al., 2011 ; Li et al., 2016 ). This east-west salinity contrast reflects the surface freshwater redistribution driven by changes in rainfall, evaporation, and ocean advection during IOD phases. To quantify these variations, we computed the DMI for the peak IOD season (SON months) and normalized it by its standard deviation. Years exceeding + 1σ were classified as pIOD events, and those below − 1σ as nIOD events (Table S1 ). Composite analysis based on this classification reveals a pronounced salinity dipole structure across the basin. During pIOD events, SSS decreases in the CEIO region, whereas SSS increases in the SJC region (Fig. S3a). Conversely, nIOD composites exhibit the opposite polarity, though the amplitude of nIOD anomalies is notably weaker (Fig. S3b-c). In addition, these findings are consistent with Li et al. ( 2016 ). Interestingly, while the spatial footprint of IOD-induced SSS anomalies remains consistent across phases, the precise definition of the salinity dipole region remains debated. A recent study by Shi & Wang ( 2025 ) proposed the SSS dipole between the conventional IOD (WIO and SEIO) regions in the IO. However, our findings suggest that the dominant salinity gradient lies between the CEIO and SJC regions, implying that the conventional SSS dipole definition may not accurately capture the true center of variability. This distinction is critical for refining salinity-based IOD indices and improving the representation of IOD-salinity interactions in climate models. To further characterize the spatial structure and coherence of the salinity dipole, we examined the correlation between DMI and SSS anomalies across the tropical IO. Consistent with previous studies indicating that the IOD-induced salinity dipole peaks during boreal autumn (Li et al., 2016 ), our analysis focused on the SON period. The correlation pattern (Fig. 2 a) closely resembles the regression structure (Fig. 1 b), showing a pronounced dipole between the CEIO and the SJC regions. Notably, the correlation is strongly negative in the CEIO, indicating significant freshening during positive IOD events, while a weaker relationship appears in the WIO. Thus, compared to the WIO region, CEIO region shows a robust response to the IOD. We further compared the correlation between the SDMI and basin-wide SSS anomalies (Fig. 2 b). The SDMI shows a relatively weak relationship over the WIO but captures stronger salinity variability in the SEIO, consistent with the regional manifestation of the salinity dipole. Therefore, based on the regression (Fig. 1 b), composite (Fig. S3), and correlation (Fig. 2 a) analyses, two representative regions were selected (CEIO and SJC) to quantify the salinity dipole characteristics (Fig. 2 c). The strongest SSS responses are concentrated in these regions, confirming their key role in the dipole dynamics. To further illustrate the out-of-phase behavior more clearly, we analyzed the correlation between the CEIO (and SJC) SSS indices and basin-wide SSS anomalies (Figs. 2 c-d). The results reveal a robust negative correlation: when SSS anomalies increase in the CEIO region, they simultaneously decrease in the SJC region, and vice versa. This reciprocal relationship provides compelling evidence for the existence of a basin-scale SSS dipole pattern in the tropical IO, driven primarily by IOD-related air-sea interaction processes. To better understand the temporal characteristics of the IOD and its associated salinity variability, we examined the amplitude and phase relationships between the IOD and salinity-based indices (Fig. 3 ). Both the DMI and newly defined SDI exhibit a dominant peak during boreal autumn, consistent with the seasonal evolution of IOD. As shown in Fig. 3 a, the DMI reaches its maximum in October (Saji et al., 1999 ), while the SDI peaks slightly later, in November. The seasonal cycles of the SDI and previously defined SDMI (Shi & Wang, 2025 ) display similar phase patterns; however, the amplitude of the SDMI is notably weaker than that of the SDI. This discrepancy likely arises from differences in the index domain, where our newly defined SDI, based on the CEIO-SJC regions, better represents the salinity response to the IOD forcing. We further explored the relationship between IOD strength and salinity variability by correlating the DMI with both the SDI and SDMI during the peak IOD season. The results clearly show a strong and statistically significant correlation between the SDI and DMI (r = − 0.87; Fig. 3 b), compared to a weaker correlation between the SDMI (defined according to Shi & Wang ( 2025 )) and DMI (r = − 0.78; Fig. 3 d). When averaged over the SON months, this relationship strengthens further, with r = − 0.93 for SDI versus r=-0.82 for SDMI, confirming that the SDI more robustly captures IOD-related salinity variability. Notably, the SDMI fails to represent the variations in SSS anomalies during several negative IOD years – particularly 1998 and 2005 (Fig. 4 d) – a limitation also noted by Grunseich et al. ( 2011 ). Moreover, during extreme negative IOD event of 2016, the SDMI exhibits only a muted signal (green circles in Fig. 3 d), whereas the SDI effectively captures the pronounced SSS anomalies associated with that event (green circles in Fig. 3 b). These findings demonstrate that the SDI provides a more robust and dynamically consistent measures of IOD-related salinity variability than the conventional SDMI, highlighting the improved sensitivity of our newly defined index to both positive and negative IOD extremes. To further support the observational results, we have analyzed the CMIP6 models and found consistent results (Fig. 5 and Fig. 6 ). 3.2. How does the IOD produce dipole variability in SSS across the Indian Ocean? To understand the physical processes underlying IOD-induced salinity variability, we examined how changes in ocean-atmosphere interactions during IOD events shape the SSS pattern (Fig. 7 ). The regression results show that the IOD explains a substantial portion of the SSS variability in the tropical IO. During positive IOD events, low-SSS anomalies dominate the CEIO, while high-SSS anomalies occur in the SJC region (Fig. 1 b and Fig. 2 a). In the SJC region, the strengthening of southeasterly winds along the coast promotes coastal upwelling (Fig. 1 a), which shoals the thermocline and reduces the thickness of the barrier layer. This vertical restructuring facilitates the entrainment of subsurface, high-salinity water into the mixed layer (Sun et al., 2022 ). The upwelled water, characterized by low temperature, high nutrients, and high salinity, contributes to surface cooling and enhances biological productivity. Consequently, the combined effects of upwelling and advection lead to increased salinity in the SJC during positive IOD events, while the opposite processes occur during negative IOD phases (Shi & Wang, 2025 ). The regression of halocline depth supports this mechanism, showing pronounced shoaling in the SJC region during positive IOD conditions (Fig. 7 d). Simultaneously, SST cooling in the SJC suppresses local convection and reduces precipitation, further contributing to salinity enhancement. Precipitations decrease markedly in the SJC during positive IOD (Fig. 7 a), whereas precipitation increases over the CEIO region, contributing to surface freshening. However, consistent with previous findings (Yuhong et al., 2013 ), changes in precipitation-evaporation balance seem to exert only a weak influence on the SJC and CEIO SSS variability. Therefore, IOD-related SSS variability in tropical IO is likely to arise from a combination of processes other than precipitation. To further examine the physical processes involved in the SSS dipole, we examined the mixed-layer salt budget. Following Li et al. ( 2016 ), four key terms – surface freshwater flux, zonal advection, meridional advection, and vertical entrainment – were considered in the salt budget framework. Since entrainment contributes marginally compared to the other terms, we focus primarily on zonal advection, meridional advection, and surface freshwater flux as the dominant processes influencing the salt budget, while entrainment is included in the residual term to explain the overall SSS dipole mechanism. The SSS anomalies typically develop during autumn, and previous studies have shown that these anomalies lag mixed-layer salinity variations by about 1–2 months (Li et al., 2016 ). Accordingly, we analyzed the mixed-layer salt budget for the August–September (AS, Fig. S5) and August–October (ASO, Fig. 8 ) periods. Figure 8 shows composites of the salt budget components averaged over five positive IOD (1994, 1997, 2006, 2015, and 2023) and six negative IOD (1996, 1998, 2005, 2010, 2016, and 2022) events during ASO. Among the horizontal advection terms, zonal advection dominates over meridional advection, indicating that zonal currents play a leading role in modulating SSS anomalies in the equatorial IO. During positive IOD events, the CEIO exhibits a negative SSS anomaly tendency primarily driven by anomalous zonal advection \(\:\left(-\left.\varvec{u}\frac{\:\partial\:\varvec{S}}{\partial\:\varvec{x}}\right)\right.\) , with a secondary contribution from anomalous meridional advection \(\:\left(\left.-\varvec{v}\frac{\partial\:\varvec{S}}{\partial\:\varvec{y}}\right)\right.\) . In contrast, the SJC shows a positive SSS anomaly tendency resulting from anomalous meridional advection and the anomalous freshwater flux \(\:\left(-\varvec{S}\frac{\left(\varvec{P}-\varvec{E}\right)}{\varvec{h}}\right)\) . During negative IOD events, a positive SSS anomaly tendency emerges in the CEIO, predominantly influenced by \(\:-\varvec{u}\frac{\:\partial\:\varvec{S}}{\partial\:\varvec{x}}\) , while \(\:-\varvec{v}\frac{\partial\:\varvec{S}}{\partial\:\varvec{y}}\) acts in the opposite direction. Meanwhile the SJC exhibits a negative SSS anomaly tendency caused by \(\:-\varvec{v}\frac{\partial\:\varvec{S}}{\partial\:\varvec{y}}\) and \(\:-\varvec{S}\frac{\left(\varvec{P}-\varvec{E}\right)}{\varvec{h}}\) . In both IOD phases, the SJC \(\:-\varvec{u}\frac{\:\partial\:\varvec{S}}{\partial\:\varvec{x}}\) term shows an opposite response relative to the observed SSS anomaly tendency. As shown in figure S5, the AS composites display a similar pattern: during positive IOD events, CEIO negative SSS anomaly tendency arises mainly from \(\:-\varvec{u}\frac{\:\partial\:\varvec{S}}{\partial\:\varvec{x}}\) with secondary contribution from \(\:-\varvec{v}\frac{\partial\:\varvec{S}}{\partial\:\varvec{y}}\) , while the SJC positive SSS anomaly tendency is driven by \(\:-\varvec{v}\frac{\partial\:\varvec{S}}{\partial\:\varvec{y}}\) and \(\:-\varvec{S}\frac{\left(\varvec{P}-\varvec{E}\right)}{\varvec{h}}\) . In contrast, during negative IOD events, positive SSS anomaly tendency in the CEIO is dominated by \(\:-\varvec{u}\frac{\:\partial\:\varvec{S}}{\partial\:\varvec{x}}\) and secondarily by \(\:-\varvec{v}\frac{\partial\:\varvec{S}}{\partial\:\varvec{y}}\) , whereas the SJC negative tendency is driven by \(\:-\varvec{v}\frac{\partial\:\varvec{S}}{\partial\:\varvec{y}}\) and \(\:-\varvec{S}\frac{\left(\varvec{P}-\varvec{E}\right)}{\varvec{h}}\) . However, \(\:-\varvec{u}\frac{\:\partial\:\varvec{S}}{\partial\:\varvec{x}}\) acts in the opposite direction against the negative SSSA tendency. Notably, the amplitude of the negative IOD SSS anomaly response is weaker than that of the positive IOD (Fig. 8 and Fig. S5). 4. Discussion and Conclusion Based on long-term observational datasets, this study investigated IOD-induced changes in SSS in the tropical IO and introduced a refined Salinity Dipole Index (SDI) that better captures these changes than the previously proposed Salinity Dipole Mode Index (SDMI). Our results reveal a robust SSS dipole during boreal Autumn, characterized by a negative SSS anomaly in the CEIO and a positive anomaly in SJC during positive IOD years, and the opposite pattern during the negative IOD years. The concept of a salinity-based IOD (SIOD) proposed by Shi & Wang, ( 2025 ) extends the framework of the conventional SST-based DMI (Saji et al., 1999 ) and the biologically derived BDMI (Shi & Wang 2021 , 2022 ). While BIOD and SIOD represent distinct aspects of the IOD, their regions differ from the conventional IOD domains and therefore require refined definitions to capture the respective IOD-related variability. For instance, the biological dipole and its related index (BDI) were defined between the south of the Indian Subcontinent and west of Sumatra (Pathirana et al., 2024 a, 2024 b; Abeywickrama et al., 2025 ). In this study, we propose a refined SDI based on CEIO–SJC regions to more accurately capture salinity variations associated with the IOD. Although a dipole pattern is evident in SSS responses to the IOD, the regions used in the SDMI (Shi and Wang, 2025 ) may not accurately represent underlying salinity variability. The conventional IOD boxes (Saji et al., 1999 ) are optimized for SST variations and therefore inadequately capture salinity signals: the western box lies too far offshore to reflect the low-salinity waters of the Bay of Bengal (BoB), while the eastern box largely reflects upwelling-related SSS anomaly in the SJC. Our refined SDI effectively captures both positive and negative IOD events and shows a stronger correlation with the DMI (r = − 0.93) than the SDMI (r = − 0.82), particularly concerning negative IODs. Notably, while the SDMI failed to capture anomalies during the negative IOD events of 1998 and 2005, the SDI successfully reproduced them. This highlights that careful regional selection enhances the sensitivity and reliability of salinity-based indices. This improvement is consistent with previous findings by Grunseich et al. ( 2011 ). Furthermore, the SDMI also underrepresents salinity variability during extreme negative IOD events, such as in 2016, highlighting the advantage of the refined SDI. Analysis of the mixed-layer salt budget demonstrates that zonal advection is the dominant mechanism governing changes in salinity in the CEIO, whereas SJC variability is primarily controlled by meridional advection and freshwater flux. A non-linear relationship is observed between positive and negative IOD phases, with stronger salinity responses during positive IOD events, suggesting nonlinear salinity feedback in the IO. The dominance of zonal advection reflects the influence of strong equatorial currents and the dynamics of equatorial Kelvin and Rossby waves. In contrast, the SJC is more affected by wind-driven coastal upwelling, precipitation and barrier layer processes. These regional contrasts highlight the spatial heterogeneity of salinity drivers within the IOD system. Our findings also emphasize the asymmetric strength of IOD-induced SSS variations: positive IODs exhibit stronger air–sea coupling and intensified easterly wind anomalies across the equator, whereas negative IODs are generally weaker and more susceptible to ENSO-induced changes in the Walker circulation (Cai & Cowan, 2013 ; Sun et al., 2022 ). The stronger amplitude of positive IOD-related salinity tendencies observed here is consistent with this asymmetric behavior. Furthermore, our salt budget analysis confirms that the magnitude of negative IOD SSS anomaly tendencies is smaller than during positive events, thus providing a dynamical explanation for the observed asymmetry. One striking feature is the contrasting SSS anomaly response observed along the eastern Indian coastline: enhanced salinity during positive IOD events and reduced salinity during negative IOD events. This pattern arises from large-scale anomalous currents modulated by zonal wind variations over the eastern IO. During negative (positive) IOD phases, anomalous westerlies (easterlies) generate downwelling (upwelling) Kelvin waves, strengthening (weakening) the Wyrtki jet and inducing basin-wide cyclonic (anticyclonic) circulation in the Bay of Bengal. These processes strengthen (weaken) the East India Coastal Current, driving the southward propagation of negative (positive) SSS anomalies along the eastern Indian coast (Chen et al., 2024 ). Together with the identified CEIO–SJC dipole, this establishes a tripole-like SSS response across the IO during the IOD peak phase. In conclusion, this study provides a comprehensive understanding of salinity variability associated with the IOD and offers a refined SDI that captures both positive and negative events with greater accuracy. By linking surface salinity changes to underlying physical processes, we emphasize the important role of advection and freshwater flux in modulating the SSS dipole. The identification of asymmetric salinity responses and the emerging tripole structure highlights the complexity of air–sea interactions in the tropical Indian Ocean. These findings improve the characterization of the IOD and provide valuable insights for the future modelling and prediction of ocean–atmosphere coupled dynamics in a changing climate. Declarations Ethics approval and consent to participate Not Applicable Consent for publication Not Applicable Conflict of interest The authors declare that they have no conflicts of interest, financial or otherwise, related to the research, authorship, and/or publication of this article. Acknowledgement We thank the World Climate Research Program and associated teams for providing the CMIP6 data. All graphs and analyses were performed using Python v. 3.12 and supportive packages. Data Availability Statement The CMIP6 data used in this study is available at https://esgf-node.llnl.gov/projects/cmip6/ . ECMWF data is available at https://www.ecmwf.int/en/forecasts/datasets . Salinity data is available at https://data.marine.copernicus.eu/products . References Abeywickrama T, Pathirana G, Noh K-M, Dissanayake K, Wang D, Lee D-G (2025) Asymmetric effects of Indian ocean dipole on surface chlorophyll variability in the Indian ocean. Ocean Dyn 75(8):69. https://doi.org/10.1007/s10236-025-01714-2 Ashok K, Guan Z, Yamagata T (2001) Impact of the Indian Ocean dipole on the relationship between the Indian monsoon rainfall and ENSO. Geophys Res Lett 28(23):4499–4502. https://doi.org/10.1029/2001GL013294 Balaguru K, Foltz GR, Leung LR, Emanuel KA (2016) Global warming-induced upper-ocean freshening and the intensification of super typhoons. Nat Commun 7(1):13670. https://doi.org/10.1038/ncomms13670 Behera SK, Luo J-J, Masson S, Delecluse P, Gualdi S, Navarra A, Yamagata T (2005) Paramount Impact of the Indian Ocean Dipole on the East African Short Rains: A CGCM Study. J Clim 18(21):4514–4530. https://doi.org/10.1175/JCLI3541.1 Bosc C, Delcroix T, Maes C (2009) Barrier layer variability in the western Pacific warm pool from 2000 to 2007. J Geophys Research: Oceans 114(C6). https://doi.org/10.1029/2008JC005187 Cai W, Cowan T (2013) Why is the amplitude of the Indian Ocean Dipole overly large in CMIP3 and CMIP5 climate models? Geophys Res Lett 40(6):1200–1205. https://doi.org/10.1002/grl.50208 Chen B, Qin H, Chen G, Xue H (2019) Ocean Salinity as a Precursor of Summer Rainfall over the East Asian Monsoon Region. J Clim 32(17):5659–5676. https://doi.org/10.1175/JCLI-D-18-0756.1 Chen S, Qiu F, Jing C, Qiu Y, Zhang J (2024) Unraveling the Influence of Equatorial Waves on Post-Monsoon Sea Surface Salinity Anomalies in the Bay of Bengal. Remote Sens 16(8):1348. https://doi.org/10.3390/rs16081348 Copernicus Marine Environment Monitoring Service (CMEMS) (2024) Global Ocean Ensemble Physics Reanalysis (1993–2023) de Boyer Montégut C, Madec G, Fischer AS, Lazar A, Iudicone D (2004) Mixed layer depth over the global ocean: An examination of profile data and a profile-based climatology. J Geophys Research: Oceans 109(C12). https://doi.org/10.1029/2004JC002378 Delcroix T, Henin C, Porte V, Arkin P (1996) Precipitation and sea-surface salinity in the tropical Pacific Ocean. Deep Sea Res Part I 43(7):1123–1141. https://doi.org/10.1016/0967-0637(96)00048-9 Doney SC, Ruckelshaus M, Emmett Duffy J, Barry JP, Chan F, English CA, Galindo HM, Grebmeier JM, Hollowed AB, Knowlton N, Polovina J, Rabalais NN, Sydeman WJ, Talley LD (2012) Climate Change Impacts on Marine Ecosystems. Annual Rev Mar Sci 4(1):11–37. https://doi.org/10.1146/annurev-marine-041911-111611 Du Y, Zhang Y (2015) Satellite and Argo Observed Surface Salinity Variations in the Tropical Indian Ocean and Their Association with the Indian Ocean Dipole Mode. J Clim 28(2):695–713. https://doi.org/10.1175/JCLI-D-14-00435.1 Grunseich G, Subrahmanyam B, Murty VSN, Giese BS (2011) Sea surface salinity variability during the Indian Ocean Dipole and ENSO events in the tropical Indian Ocean. J Geophys Research: Oceans 116(C11). https://doi.org/10.1029/2011JC007456 Hersbach H (2023) B. B. ; B. P. ; B. G. ; H. A. ; M. S. J. ; N. J. ; P. C. ; R. R. ; R. I. ; S. D. ; S. A. ; S. C. ; D. D. ; T. J.-N. ERA5 monthly averaged data on single levels from 1940 to present . https://doi.org/10.24381/cds.f17050d7 Hu S, Zhang Y, Feng M, Du Y, Sprintall J, Wang F, Hu D, Xie Q, Chai F (2019) Interannual to Decadal Variability of Upper-Ocean Salinity in the Southern Indian Ocean and the Role of the Indonesian Throughflow. J Clim 32(19):6403–6421. https://doi.org/10.1175/JCLI-D-19-0056.1 Jensen TG (2001) Arabian Sea and Bay of Bengal exchange of salt and tracers in an ocean model. Geophys Res Lett 28(20):3967–3970. https://doi.org/10.1029/2001GL013422 Jinadasa SUP, Pathirana G, Ranasinghe PN, Centurioni L, Hormann V (2020) Monsoonal impact on circulation pathways in the Indian Ocean. Acta Oceanol Sin 39(3):103–112. https://doi.org/10.1007/s13131-020-1557-5 Kumar SP, Prasad TG (1999) Formation and spreading of Arabian Sea high-salinity water mass. J Geophys Research: Oceans 104(C1):1455–1464. https://doi.org/10.1029/1998JC900022 Lagerloef GSE (2002) Introduction to the special section: The role of surface salinity on upper ocean dynamics, air-sea interaction and climate. J Geophys Research: Oceans 107(C12). https://doi.org/10.1029/2002JC001669 Li J, Liang C, Tang Y, Dong C, Chen D, Liu X, Jin W (2016) A new dipole index of the salinity anomalies of the tropical Indian Ocean. Sci Rep 6(1):24260. https://doi.org/10.1038/srep24260 Li Z, Lian T, Ying J, Zhu X, Papa F, Xie H, Long Y (2021) The Cause of an Extremely Low Salinity Anomaly in the Bay of Bengal During 2012 Spring. J Geophys Research: Oceans 126(10). https://doi.org/10.1029/2021JC017361 Pathirana G, Min Noh K, Lee D-G, Park H-J, Wang D, Kug J-S (2024) Intensified surface chlorophyll responses to the Indian Ocean Dipole under greenhouse warming. Environ Res Lett 19(6):064072. https://doi.org/10.1088/1748-9326/ad4db8 Pathirana G, Noh KM, Lee D-G, Lee H, Kug J-S (2024) A biological dipole variability in the Indian Ocean. Environ Res Lett 19(1):014070. https://doi.org/10.1088/1748-9326/ad1641 Perigaud C, McCreary JP, Zhang KQ (2003) Impact of interannual rainfall anomalies on Indian Ocean salinity and temperature variability. J Geophys Research: Oceans 108(C10). https://doi.org/10.1029/2002JC001699 Rao RR, Kumar KVS (1991) Evolution of salinity field in the upper layers of the east central Arabian Sea and northern Bay of Bengal during summer monsoon experiments. J Earth Syst Sci 100(1):69–78. https://doi.org/10.1007/BF02843485 Rao RR, Sivakumar R (2003) Seasonal variability of sea surface salinity and salt budget of the mixed layer of the north Indian Ocean. J Geophys Research: Oceans 108(C1). https://doi.org/10.1029/2001JC000907 Saji NH, Goswami BN, Vinayachandran PN, Yamagata T (1999) A dipole mode in the tropical Indian Ocean. Nature 401(6751):360–363. https://doi.org/10.1038/43854 Saji N, Yamagata T (2003) Possible impacts of Indian Ocean Dipole mode events on global climate. Climate Res 25:151–169. https://doi.org/10.3354/cr025151 Sandeep KK, Pant V, Girishkumar MS, Rao AD (2018) Impact of riverine freshwater forcing on the sea surface salinity simulations in the Indian Ocean. J Mar Syst 185:40–58. https://doi.org/10.1016/j.jmarsys.2018.05.002 Schott FA, McCreary JP (2001) The monsoon circulation of the Indian Ocean. Prog Oceanogr 51(1):1–123. https://doi.org/10.1016/S0079-6611(01)00083-0 Shi W, Wang M (2021) A biological Indian Ocean Dipole event in 2019. Sci Rep 11(1):2452. https://doi.org/10.1038/s41598-021-81410-5 Shi W, Wang M (2022) Biological dipole mode indices: New parameters to characterize the physical and biological processes of the Indian Ocean Dipole event. Prog Oceanogr 206:102847. https://doi.org/10.1016/j.pocean.2022.102847 Shi W, Wang M (2025) Salinity Indian Ocean Dipole: Another facet of the Indian Ocean Dipole phenomenon from satellite remote sensing. Sci Remote Sens 11:100184. https://doi.org/10.1016/j.srs.2024.100184 Stammer D, Martins MS, Köhler J, Köhl A (2021) How well do we know ocean salinity and its changes? Prog Oceanogr 190:102478. https://doi.org/10.1016/j.pocean.2020.102478 Subrahmanyam B, Murty VSN, Heffner DM (2011) Sea surface salinity variability in the tropical Indian Ocean. Remote Sens Environ 115(3):944–956. https://doi.org/10.1016/j.rse.2010.12.004 Sun Q, Zhang Y, Du Y, Jiang X (2022) Asymmetric Response of Sea Surface Salinity to Extreme Positive and Negative Indian Ocean Dipole in the Southern Tropical Indian Ocean. J Geophys Research: Oceans 127(11). https://doi.org/10.1029/2022JC018986 Thompson B, Gnanaseelan C, Salvekar PS (2006) Variability in the Indian Ocean circulation and salinity and its impact on SST anomalies during dipole events. J Mar Res 64(6):853–880. https://doi.org/10.1357/002224006779698350 Wyrtki K (1973) An Equatorial Jet in the Indian Ocean. Science 181(4096):262–264. https://doi.org/10.1126/science.181.4096.262 Yuan C, Yamagata T (2015) Impacts of IOD, ENSO and ENSO Modoki on the Australian Winter Wheat Yields in Recent Decades. Sci Rep 5(1):17252. https://doi.org/10.1038/srep17252 Yuhong Z, Yan D, Shaojun Z, Yali Y, Xuhua C (2013) Impact of Indian Ocean Dipole on the salinity budget in the equatorial Indian Ocean. J Geophys Research: Oceans 118(10):4911–4923. https://doi.org/10.1002/jgrc.20392 Zhang Y, Du Y, Qu T (2016) A sea surface salinity dipole mode in the tropical Indian Ocean. Clim Dyn 47(7–8):2573–2585. https://doi.org/10.1007/s00382-016-2984-z Supplementary Files SupplementarySDI.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviewers agreed at journal 03 Feb, 2026 Reviewers invited by journal 02 Feb, 2026 Editor assigned by journal 22 Dec, 2025 First submitted to journal 18 Dec, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8401709","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":584621627,"identity":"332f2b12-ce2b-460e-805c-1ea602cbe9c5","order_by":0,"name":"Kokulathasan Thilaksanth","email":"","orcid":"","institution":"University of Ruhuna","correspondingAuthor":false,"prefix":"","firstName":"Kokulathasan","middleName":"","lastName":"Thilaksanth","suffix":""},{"id":584621628,"identity":"20e4262f-8a8c-4c7b-a986-6b67e66bf977","order_by":1,"name":"Gayan Pathirana","email":"data:image/png;base64,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","orcid":"https://orcid.org/0000-0002-4670-579X","institution":"University of Ruhuna","correspondingAuthor":true,"prefix":"","firstName":"Gayan","middleName":"","lastName":"Pathirana","suffix":""},{"id":584621629,"identity":"7cd19ee3-a85f-4eb3-b065-614ca3b518e5","order_by":2,"name":"Kyung-Min Noh","email":"","orcid":"","institution":"Princeton University","correspondingAuthor":false,"prefix":"","firstName":"Kyung-Min","middleName":"","lastName":"Noh","suffix":""},{"id":584621630,"identity":"804882d5-f6bf-4c7e-b02a-85f47660342f","order_by":3,"name":"Aoyun Xue","email":"","orcid":"","institution":"University of California Santa Barbara","correspondingAuthor":false,"prefix":"","firstName":"Aoyun","middleName":"","lastName":"Xue","suffix":""},{"id":584621631,"identity":"28198bdd-8549-46f9-984d-b9563bd86b61","order_by":4,"name":"Dongxiao Wang","email":"","orcid":"","institution":"Sun Yat-Sen University","correspondingAuthor":false,"prefix":"","firstName":"Dongxiao","middleName":"","lastName":"Wang","suffix":""},{"id":584621632,"identity":"3d98d7f0-9ea6-456c-a456-46efc71c5e30","order_by":5,"name":"Zheng Meng","email":"","orcid":"","institution":"Sun Yat-Sen University","correspondingAuthor":false,"prefix":"","firstName":"Zheng","middleName":"","lastName":"Meng","suffix":""}],"badges":[],"createdAt":"2025-12-19 07:09:47","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8401709/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8401709/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101844460,"identity":"57801b95-de43-4bd5-9f25-7c50b93e6352","added_by":"auto","created_at":"2026-02-04 09:07:29","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":853716,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Regressed SST and winds on DMI. (b) Regressed SSS on DMI. for observations from 1993-2023. The chosen period is September to November. (a) Wind vectors significant at the 95% confidence level are shown in green, while non-significant vectors are shown in yellow. And (b). Significant regions at the 95% confidence interval are marked by contours.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-8401709/v1/b694cfe9d6b0ac01743237dd.png"},{"id":101844455,"identity":"8c383410-724e-4bd1-abf9-16764b114fd5","added_by":"auto","created_at":"2026-02-04 09:07:29","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":711670,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Correlation between DMI and surface salinity anomalies. (b) Correlation between SDMI and surface salinity anomalies. In (a) and (b) the black solid and dashed square indicates the southeastern IO (SEIO) and western IO (WIO) region respectively, used to define the DMI. (c) Correlation between CEIO salinity index and surface salinity anomalies, and (d) correlation between SJC salinity index and surface salinity anomalies. In (c) and (d) the purple solid and dashed square indicates the SJC and CEIO regions used to define the salinity dipole index (SDI). In each plot significant regions at the 95% confidence interval are marked by contours.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-8401709/v1/a7eda65eace6769f6c62b065.png"},{"id":101880812,"identity":"c312505a-edb2-4bbc-a77e-9ff2a9d74819","added_by":"auto","created_at":"2026-02-04 15:06:37","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":112315,"visible":true,"origin":"","legend":"\u003cp\u003eObserved interannual variability of (a) DMI (orange) and SDI (blue) during 1993–2023 and observed (b) correlation between DMI and SDI during 1993–2023 for September to November. (c) and (d), Same as (a) and (b), respectively, but for observed DMI and SDMI. The regressed line is shown in red, and the correlation (r) shown in the plot is significant at the 99% confidence level. In (b) and (d), the October and November months of 2016 extreme negative and the 1997 extreme positive IOD events are marked using green labels.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-8401709/v1/728f46433f43abe46dcc351b.png"},{"id":101881521,"identity":"408f2ec8-5407-4758-8e41-5f4281e95f35","added_by":"auto","created_at":"2026-02-04 15:12:55","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":152056,"visible":true,"origin":"","legend":"\u003cp\u003eTime series of (a) DMI (black) and SDI (green) during 1993–2023 and (b) their correlation for SON. Panels (c) and (d) are the same as (a) and (b), respectively, but for SDMI (blue). The regression line is shown in green, with r values significant at the 99% confidence level. In (b) and (d), red, blue, and black circles mark extreme positive, extreme negative, and normal IOD years, respectively, during SON.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-8401709/v1/14a62c3774ca8a3480241a75.png"},{"id":101881531,"identity":"5bfa7bbb-eab6-457b-ad5b-7b660acec54c","added_by":"auto","created_at":"2026-02-04 15:13:02","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":78801,"visible":true,"origin":"","legend":"\u003cp\u003ePattern correlation coefficients (PCC) between observation and 13 CMIP6 model SSS. Red bars show the mean SSS PCC, blue bars show the interannual variability PCC. The green dashed line (PCC = 0.5) indicates the performance threshold.\u003c/p\u003e","description":"","filename":"5.png","url":"https://assets-eu.researchsquare.com/files/rs-8401709/v1/4a9d540fdc2ff6daf1011e32.png"},{"id":101881946,"identity":"0293d9c7-ca52-4c2e-997d-285f907627a2","added_by":"auto","created_at":"2026-02-04 15:17:45","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":109471,"visible":true,"origin":"","legend":"\u003cp\u003eCorrelation between DMI and SDI (blue) and SDMI (lime green) for observations and 13 CMIP6 models during SON for the period 1850–2014. The bars represent correlation strength, with observational results shown first.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-8401709/v1/05e4ef1b8733b1a274e24511.png"},{"id":101844463,"identity":"4a2c9a60-498d-4376-882d-c2c0fb1d3276","added_by":"auto","created_at":"2026-02-04 09:07:30","extension":"png","order_by":7,"title":"Figure 7","display":"","copyAsset":false,"role":"figure","size":824730,"visible":true,"origin":"","legend":"\u003cp\u003e(a) Regressed Rainfall on DMI, (b) Regressed Evaporation on DMI, (c) Regressed Sea Surface Current on DMI, and (d) Regressed Halocline on DMI. for observations from 1993-2023. The chosen period is September to November. Significant regions at the 95% confidence interval are marked by contours.\u003c/p\u003e","description":"","filename":"7.png","url":"https://assets-eu.researchsquare.com/files/rs-8401709/v1/2d9b26019f718a91fbd5dbff.png"},{"id":101844457,"identity":"3521638e-f7f9-4e60-9efd-15378d5141ab","added_by":"auto","created_at":"2026-02-04 09:07:29","extension":"png","order_by":8,"title":"Figure 8","display":"","copyAsset":false,"role":"figure","size":91181,"visible":true,"origin":"","legend":"\u003cp\u003eComposited salinity budget components averaged over August–September during IOD events. (a, b) show salinity budget components averaged over the CEIO and SJC regions for pIOD events, respectively. (c, d) are the same as (a, b) but for nIOD events.\u003c/p\u003e","description":"","filename":"8.png","url":"https://assets-eu.researchsquare.com/files/rs-8401709/v1/299758fb5137d593a7f94cca.png"},{"id":101943702,"identity":"b66111e2-064e-4e23-b553-eb9627e3e68e","added_by":"auto","created_at":"2026-02-05 09:42:55","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3223282,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8401709/v1/0795bd7f-24ea-45c7-b2f2-d2e5300ef0e6.pdf"},{"id":101844462,"identity":"6f955c25-bbf0-43c5-99f7-3c6c642ed7e0","added_by":"auto","created_at":"2026-02-04 09:07:30","extension":"docx","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":1922983,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementarySDI.docx","url":"https://assets-eu.researchsquare.com/files/rs-8401709/v1/29ad43f5d9396c6c73ac8c98.docx"}],"financialInterests":"","formattedTitle":"A Refined Perspective on Indian Ocean Dipole–Driven Surface Salinity Changes and the Salinity Dipole Index","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eOcean salinity is a fundamental physical property of the ocean that plays a crucial role in large-scale circulation, air\u0026ndash;sea interactions, and climate variability (Stammer et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Salinity influences thermohaline circulation (Hu et al., \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2019\u003c/span\u003e) and ocean stratification (Balaguru et al., \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), acting as a key regulator of global climate systems. Variations in salinity further impact marine ecosystems by altering ocean productivity, food web dynamics, and marine species distribution (Doney et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2012\u003c/span\u003e). At a regional scale, surface salinity has a strong influence on the thickness of barrier layers (Bosc et al., \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2009\u003c/span\u003e) and the depth of ocean mixed layers (de Boyer Mont\u0026eacute;gut et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2004\u003c/span\u003e), as well as on subduction processes, thereby linking ocean thermodynamics with the hydrological cycle (Lagerloef, \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e2002\u003c/span\u003e). Further, sea surface salinity (SSS) reflects the combined influence of freshwater fluxes (evaporation, precipitation, and river discharge), ocean advection, entrainment, and vertical mixing (Delcroix et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e1996\u003c/span\u003e; Perigaud et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Sandeep et al., \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). SSS therefore represents both a driver of ocean dynamics and a sensitive indicator of freshwater fluxes, ocean circulation, and climate variability.\u003c/p\u003e \u003cp\u003eThe Indian Ocean (IO) is one of the most hydrographically diverse basins, with a SSS distribution that reflects strong regional contrasts. In particular, the Arabian Sea (AS) in the northern IO is characterized by high salinity due to net evaporation (Kumar \u0026amp; Prasad, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1999\u003c/span\u003e), while the Bay of Bengal (BoB) remains persistently fresh due to river discharge and high precipitation (Rao \u0026amp; Kumar, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e1991\u003c/span\u003e). On a seasonal timescale, the monsoon seasons have a significant influence on the spatial variability of surface winds and currents in the northern IO (Jinadasa et al., \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Schott \u0026amp; McCreary, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). In response to the shift in winds and surface currents from the southwest monsoon (SWM) to the northeast monsoon (NEM), water is exchanged between the AS and BoB basins surrounding Sri Lanka. This reflects a mixture of low and high salinity waters on a seasonal timescale. In contrast, the Central Equatorial Indian Ocean (CEIO) exhibits a semi-annual salinity cycle linked to the monsoon system (Yuhong et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). In addition to these seasonal and regional processes, the interannual and decadal variability of SSS in the IO is significantly influenced by coupled climate modes, particularly the Indian Ocean Dipole (IOD) and the El Ni\u0026ntilde;o\u0026ndash;Southern Oscillation (ENSO) (Du \u0026amp; Zhang, \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Jensen, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Rao \u0026amp; Sivakumar, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Saji \u0026amp; Yamagata, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2003\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eAmong the tropical air-sea coupled climate modes, the IOD is the dominant interannual mode of climate variability in the tropical IO. It is defined by opposing sea surface temperature anomalies (SSTAs) between the western (10\u0026deg; S\u0026ndash;10\u0026deg; N, 50\u0026deg; E\u0026ndash;70\u0026deg; E) and southeastern equatorial (10\u0026deg; S\u0026ndash;0\u0026deg;, 90\u0026deg; E\u0026ndash;110\u0026deg; E) regions (Saji et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). Developing during the boreal summer and peaking in the autumn, the IOD influences rainfall, agriculture, fisheries, and even human health across the IO-rim countries and beyond (Ashok et al., \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2001\u003c/span\u003e; Behera et al., \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Yuan \u0026amp; Yamagata, \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2015\u003c/span\u003e). Notably, IOD events are associated not only with SST anomalies, but also with pronounced salinity anomalies (SSSA), which in turn feed back into the coupled system, modulating IOD evolution (Rao \u0026amp; Sivakumar, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2003\u003c/span\u003e). Several studies have highlighted robust salinity signatures during IOD phases. Positive IOD (pIOD) events are characterized by freshening in the CEIO and salinification along the Sumatra\u0026ndash;Java Coast (SJC), whereas negative IOD (nIOD) events exhibit opposite patterns (Li et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; Subrahmanyam et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Thompson et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). These contrasting anomalies form a distinct salinity dipole. To quantify this phenomenon,Li et al. (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2016\u003c/span\u003e) introduced the Dipole Mode Index of Salinity (DMIS), which is defined as the difference in SSSA between the CEIO and the SJC. More recently,Shi \u0026amp; Wang, (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) introduced a novel salinity dipole mode index (SDMI) based on the conventional SST-defined IOD regions. However, this latter approach assumes that salinity variability mirrors SST anomalies spatially, a premise that has been challenged by observational and modelling studies alike.\u003c/p\u003e \u003cp\u003eSuch inconsistencies in defining the salinity dipole index are like debates surrounding other dipole indices. For instance, Shi \u0026amp; Wang (\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2021\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2022\u003c/span\u003e) introduced a biological IOD (BIOD) and a related index based on conventional IOD regions. However, subsequent research demonstrated that biological responses were more accurately represented by alternative regions, resulting in the development of a refined biological dipole index (BDI) (Pathirana et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2024\u003c/span\u003ea, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2024\u003c/span\u003eb; Abeywickrama et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). This suggests that the accuracy of dipole indices in representing the underlying variability is critically determined by the choice of index regions. In the case of SSS, it remains an open question whether the dipole is best captured by the CEIO\u0026ndash;SJC framework or by conventional IOD regions. Against this backdrop, revisiting the definition of a salinity dipole index is essential. A robust, physically consistent Salinity Dipole Index (SDI) would advance our understanding of salinity variability during IOD events and improve the representation of freshwater fluxes, surface circulation, and air\u0026ndash;sea coupling in the tropical IO. Furthermore, refining the SDI has implications for climate prediction, given that SSS is increasingly recognized as a key precursor and modulator of ocean\u0026ndash;atmosphere feedback (Chen et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2019\u003c/span\u003e). In this study, we re-evaluate the relationship between the IOD and SSS using long-term observational and historical simulations from a Coupled Model Intercomparison Project Phase 6 (CMIP6) model datasets, proposing a refined formulation of the salinity dipole regions and index. This refined SDI provides a clearer and more consistent measure of salinity variability, offering a critical tool for both fundamental climate research and applied prediction efforts.\u003c/p\u003e"},{"header":"2. Data and Methods","content":"\u003cp\u003eTo investigate the influence of the IOD on SSS and to refine the salinity dipole index (SDI) definition, we analyzed oceanic and atmospheric datasets from 1993 to 2023. We obtained monthly SST, SSS, surface current, and subsurface temperature and salinity from the Copernicus Marine Environment Monitoring Service (CMEMS) reanalysis products (CMEMS, 2024). Atmospheric variables, including 10 m winds, precipitation, and evaporation, were obtained from the ERA5 reanalysis, produced by the European Centre for Medium-Range Weather Forecasts (ECMWF; Hersbach, \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). In addition to observational data, historical simulations from the CMIP6 were analyzed using 13 CMIP6 models selected based on data availability (Table S2). Details of each dataset, including their spatial resolution, temporal coverage and access links, are summarized in Supplementary Table\u0026nbsp;3. To isolate variability, we calculated monthly anomalies by removing the mean seasonal cycle from each field.\u003c/p\u003e \u003cp\u003eOur analysis proceeded in three stages. First, we examined large-scale salinity variability in relation to the dominant climate modes. We then applied multiple linear regression to regress SSS and SST anomalies separately onto the Dipole Mode Index (DMI; September\u0026ndash;November, SON) and the Ni\u0026ntilde;o 3.4 index (December\u0026ndash;February, DJF). This enabled us to determine the relative influence of IOD and ENSO on IO salinity variability. Based on these results, subsequent analyses focused on the SON season, when IOD impacts are strongest. For the SON season, we computed the SSS climatology and interannual standard deviation to characterize the baseline state and variability. Positive and negative IOD years were defined as those with normalized DMI values exceeding\u0026thinsp;\u0026plusmn;\u0026thinsp;1σ (Li et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Composite SSS anomalies were then constructed to assess dipole patterns associated with pIOD and nIOD events.\u003c/p\u003e \u003cp\u003eSecond, we compared alternative definitions of the salinity dipole. Time series of SSS anomalies were extracted from the Central Equatorial Indian Ocean (CEIO: 70\u0026deg;E\u0026ndash;90\u0026deg;E, 5\u0026deg;S\u0026ndash;5\u0026deg;N) and the Sumatra\u0026ndash;Java Coast (SJC: 100\u0026deg;E\u0026ndash;110\u0026deg;E, 13\u0026deg;S\u0026ndash;3\u0026deg;S) regions. These were then used to compute a CEIO\u0026ndash;SJC salinity index. This was then contrasted with the Salinity Dipole Mode Index (SDMI) proposed by Shi and Wang (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2025\u003c/span\u003e), which follows the conventional IOD regions. Correlation analyses were conducted between DMI, SDMI and CEIO\u0026ndash;SJC\u0026ndash;based SDI to evaluate which formulation more robustly captures salinity variability during IOD phases.\u003c/p\u003e \u003cp\u003eThirdly, we investigated the physical mechanisms driving the salinity dipole. Multiple regression analyses were performed on SON SSS anomalies in relation to key forcing fields, such as precipitation, evaporation, surface currents, and halocline depth. Halocline depth was estimated as the depth of the maximum vertical salinity gradient, computed from subsurface salinity profiles. To further quantify the relative roles of surface fluxes and ocean dynamics, we performed a mixed-layer salt budget analysis in accordance with the methodology of Li et al. (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). The salt budget equation is expressed as follows (Sun et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2022\u003c/span\u003e):\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$\\:\\frac{\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{t}}=\\:-\\left(\\left.\\varvec{u}\\frac{\\:\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{x}}+\\varvec{v}\\frac{\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{y}}\\right)\\right.-\\varvec{S}\\frac{\\left(\\varvec{P}-\\varvec{E}\\right)}{\\varvec{h}}+\\:\\varvec{\\epsilon\\:}$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eWhere \u003cb\u003eS\u003c/b\u003e is the mixed-layer salinity, \u003cb\u003et\u003c/b\u003e is time, \u003cb\u003eu\u003c/b\u003e and \u003cb\u003ev\u003c/b\u003e are the zonal and meridional velocity components, \u003cb\u003eP\u003c/b\u003e and \u003cb\u003eE\u003c/b\u003e are the precipitation and evaporation, and \u003cb\u003eh\u003c/b\u003e is the mixed-layer depth, and (\u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\varvec{\\epsilon\\:}\\)\u003c/span\u003e\u003c/span\u003e) is the residual term representing unresolved processes. The mixed-layer depth was defined using the temperature criterion (de Boyer Mont\u0026eacute;gut et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2004\u003c/span\u003e). A summary of index definitions, acronyms, and regional boundaries used in this study is summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e for reference.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eInformation on index regions, definitions, and acronyms used in the present study.\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"2\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eIO\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIndian Ocean\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eIOD\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eIndian Ocean Dipole\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eDMI\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eDipole Mode Index. DMI is defined as the difference in SST anomalies between the western Indian Ocean (10\u0026deg; S\u0026ndash;10\u0026deg; N, 50\u0026deg; E\u0026ndash;70\u0026deg; E) and the southeastern Indian Ocean (10\u0026deg; S\u0026ndash;0\u0026deg;, 90\u0026deg; E\u0026ndash;110\u0026deg; E, Saji et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1999\u003c/span\u003e).\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eWIO\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eWestern Indian Ocean (10\u003csup\u003e0\u003c/sup\u003e S-10\u003csup\u003e0\u003c/sup\u003e N, 50\u003csup\u003e0\u003c/sup\u003e E-70\u003csup\u003e0\u003c/sup\u003e E)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSEIO\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSoutheastern Indian Ocean (10\u003csup\u003e0\u003c/sup\u003e S-0\u003csup\u003e0\u003c/sup\u003e, 90\u003csup\u003e0\u003c/sup\u003e E-110\u003csup\u003e0\u003c/sup\u003e E)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSDMI\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSalinity Dipole Mode Index (Shi \u0026amp; Wang, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). SDMI is defined as the difference of average SSS anomaly in the Western Indian Ocean (10\u003csup\u003e0\u003c/sup\u003e S-10\u003csup\u003e0\u003c/sup\u003e N, 50\u003csup\u003e0\u003c/sup\u003e E-70\u003csup\u003e0\u003c/sup\u003e E) and the Southeastern Indian Ocean (10\u003csup\u003e0\u003c/sup\u003e S-0\u003csup\u003e0\u003c/sup\u003e, 90\u003csup\u003e0\u003c/sup\u003e E-110\u003csup\u003e0\u003c/sup\u003e E).\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eCEIO\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCentral Equatorial Indian Ocean (5\u003csup\u003e0\u003c/sup\u003eS\u0026ndash;5\u003csup\u003e0\u003c/sup\u003eN, 70\u003csup\u003e0\u003c/sup\u003eE\u0026ndash;90\u003csup\u003e0\u003c/sup\u003eE)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSJC\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eSumatra Java Coast (13\u003csup\u003e0\u003c/sup\u003eS\u0026ndash;3\u003csup\u003e0\u003c/sup\u003eN, 100\u003csup\u003e0\u003c/sup\u003eE\u0026ndash;110\u003csup\u003e0\u003c/sup\u003eE)\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cb\u003eSDI*\u003c/b\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eA salinity dipole index based on the SSS anomalies between the central equatorial Indian Ocean (5\u003csup\u003e0\u003c/sup\u003eS\u0026ndash;5\u003csup\u003e0\u003c/sup\u003eN, 70\u003csup\u003e0\u003c/sup\u003eE\u0026ndash;90\u003csup\u003e0\u003c/sup\u003eE) and the region off the Sumatra\u0026ndash;Java coast (13\u003csup\u003e0\u003c/sup\u003eS\u0026ndash;3\u003csup\u003e0\u003c/sup\u003eN, 100\u003csup\u003e0\u003c/sup\u003eE\u0026ndash;110\u003csup\u003e0\u003c/sup\u003eE) was first proposed as the Dipole Mode Index for Salinity (DMIS) by (Li et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2016\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eIn the present study, we introduce a refined Salinity Dipole Index (SDI), to more clearly quantify the SSS dipole pattern associated with the IOD.\u003c/p\u003e \u003cp\u003eThe SDI is defined as the difference between the average SSS anomaly over the CEIO and that over the SJC region.\u003c/p\u003e \u003cp\u003eCompared to the previously defined SDMI, the SDI more directly captures the opposing salinity anomalies in the CEIO and SJC.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e"},{"header":"3. Results","content":"\u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e3.1. Dipole pattern in Sea Surface Salinity\u003c/h2\u003e \u003cp\u003eDuring the IOD peak season, the mean state of SSS shows clear regional differences north of 20\u0026deg;S, reflecting the combined effects of air\u0026ndash;sea fluxes, river discharge, and ocean circulation (see Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ea). The Bay of Bengal (BoB) and the eastern Indian Ocean (EIO) have relatively fresh surface waters, mainly due to more precipitation than evaporation and a lot of freshwaters coming from major rivers (Rao \u0026amp; Kumar, \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e1991\u003c/span\u003e; Li et al., \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). In contrast, the Arabian Sea (AS) has high salinity due to persistent net evaporation (Kumar \u0026amp; Prasad, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). This large-scale salinity gradient highlights the complex interaction between freshwater fluxes, ocean mixing, and circulation processes. Notably, a high-salinity tongue extends eastward along the equator (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ea), formed by the eastward advection of saline waters by the equatorial jet (Wyrtki, \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e1973\u003c/span\u003e). Meanwhile, a low-salinity tongue near 12\u0026deg;S (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ea) originates from the westward advection of fresher waters by the South Equatorial Current. These current transport low-salinity waters from the EIO and the Indonesian Throughflow (Yuhong et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). The interannual variability of SSS, represented by its standard deviation (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eb), reveals distinct spatial patterns. The northern BoB exhibits strong variability, primarily due to variable river runoff, while regions along the east coast of India, the Andaman Sea, and the central equatorial Indian Ocean (CEIO) demonstrate pronounced fluctuations linked to large-scale climate phenomena such as the Indian Ocean Dipole (IOD) and the El Ni\u0026ntilde;o\u0026ndash;Southern Oscillation (ENSO) (Chen et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). These patterns emphasize the dynamic nature of surface salinity and its sensitivity to local hydrological inputs and basin-scale climatic variability.\u003c/p\u003e \u003cp\u003eSince IOD and ENSO are the major climate modes influencing SSS in tropical IO, we applied a multiple linear regression analysis to isolate their respective impacts. The regression patterns representing IOD-driven variations in SST, surface circulation, and SSS during the peak IOD phase in boreal autumn are presented in Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e, while the ENSO-related effects are shown separately in Fig. S2. In this paper, we focus primarily on the IOD-induced changes, as our regression framework effectively removes ENSO-induced changes to reveal the independent IOD signal. The IOD regression pattern clearly depicts the canonical east-west dipole structure, with significant cooling in the SEIO and warming in the WIO (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea). These temperature anomalies are accompanied by intensified southeasterly winds, driven by the SST-induced pressure gradient force. The resulting surface wind anomalies promote enhanced upwelling and mixing in the SEIO and downwelling in the WIO, further amplifying the SST contrast. The strengthened easterly winds transport cooler, drier air from the SEIO toward the warmer, convectively active WIO, reinforcing the thermodynamic and dynamic coupling between ocean and atmosphere. This feedback loop, characteristic of the positive IOD phase, exemplifies the development of the Bjerknes feedback mechanism that sustains and amplifies IOD events.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo further assess how the IOD modulates SSS across tropical IO, we examined the regression patterns of SSS anomalies associated with IOD (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb). During pIOD events, SSS anomalies increase markedly in the SJC and eastern Indian coastal regions, while negative anomalies dominate the CEIO. The opposite pattern emerges during nIOD events, consistent with previous studies (Thompson et al., \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Subrahmanyam et al., \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Li et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). This east-west salinity contrast reflects the surface freshwater redistribution driven by changes in rainfall, evaporation, and ocean advection during IOD phases. To quantify these variations, we computed the DMI for the peak IOD season (SON months) and normalized it by its standard deviation. Years exceeding\u0026thinsp;+\u0026thinsp;1σ were classified as pIOD events, and those below \u0026minus;\u0026thinsp;1σ as nIOD events (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). Composite analysis based on this classification reveals a pronounced salinity dipole structure across the basin. During pIOD events, SSS decreases in the CEIO region, whereas SSS increases in the SJC region (Fig. S3a). Conversely, nIOD composites exhibit the opposite polarity, though the amplitude of nIOD anomalies is notably weaker (Fig. S3b-c). In addition, these findings are consistent with Li et al. (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Interestingly, while the spatial footprint of IOD-induced SSS anomalies remains consistent across phases, the precise definition of the salinity dipole region remains debated. A recent study by Shi \u0026amp; Wang (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) proposed the SSS dipole between the conventional IOD (WIO and SEIO) regions in the IO. However, our findings suggest that the dominant salinity gradient lies between the CEIO and SJC regions, implying that the conventional SSS dipole definition may not accurately capture the true center of variability. This distinction is critical for refining salinity-based IOD indices and improving the representation of IOD-salinity interactions in climate models.\u003c/p\u003e \u003cp\u003eTo further characterize the spatial structure and coherence of the salinity dipole, we examined the correlation between DMI and SSS anomalies across the tropical IO. Consistent with previous studies indicating that the IOD-induced salinity dipole peaks during boreal autumn (Li et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), our analysis focused on the SON period. The correlation pattern (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea) closely resembles the regression structure (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb), showing a pronounced dipole between the CEIO and the SJC regions. Notably, the correlation is strongly negative in the CEIO, indicating significant freshening during positive IOD events, while a weaker relationship appears in the WIO. Thus, compared to the WIO region, CEIO region shows a robust response to the IOD. We further compared the correlation between the SDMI and basin-wide SSS anomalies (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb). The SDMI shows a relatively weak relationship over the WIO but captures stronger salinity variability in the SEIO, consistent with the regional manifestation of the salinity dipole. Therefore, based on the regression (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb), composite (Fig. S3), and correlation (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea) analyses, two representative regions were selected (CEIO and SJC) to quantify the salinity dipole characteristics (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec). The strongest SSS responses are concentrated in these regions, confirming their key role in the dipole dynamics. To further illustrate the out-of-phase behavior more clearly, we analyzed the correlation between the CEIO (and SJC) SSS indices and basin-wide SSS anomalies (Figs.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ec-d). The results reveal a robust negative correlation: when SSS anomalies increase in the CEIO region, they simultaneously decrease in the SJC region, and vice versa. This reciprocal relationship provides compelling evidence for the existence of a basin-scale SSS dipole pattern in the tropical IO, driven primarily by IOD-related air-sea interaction processes.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo better understand the temporal characteristics of the IOD and its associated salinity variability, we examined the amplitude and phase relationships between the IOD and salinity-based indices (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Both the DMI and newly defined SDI exhibit a dominant peak during boreal autumn, consistent with the seasonal evolution of IOD. As shown in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea, the DMI reaches its maximum in October (Saji et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1999\u003c/span\u003e), while the SDI peaks slightly later, in November. The seasonal cycles of the SDI and previously defined SDMI (Shi \u0026amp; Wang, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) display similar phase patterns; however, the amplitude of the SDMI is notably weaker than that of the SDI. This discrepancy likely arises from differences in the index domain, where our newly defined SDI, based on the CEIO-SJC regions, better represents the salinity response to the IOD forcing. We further explored the relationship between IOD strength and salinity variability by correlating the DMI with both the SDI and SDMI during the peak IOD season. The results clearly show a strong and statistically significant correlation between the SDI and DMI (r = \u0026minus;\u0026thinsp;0.87; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb), compared to a weaker correlation between the SDMI (defined according to Shi \u0026amp; Wang (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2025\u003c/span\u003e)) and DMI (r = \u0026minus;\u0026thinsp;0.78; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ed). When averaged over the SON months, this relationship strengthens further, with r = \u0026minus;\u0026thinsp;0.93 for SDI versus r=-0.82 for SDMI, confirming that the SDI more robustly captures IOD-related salinity variability. Notably, the SDMI fails to represent the variations in SSS anomalies during several negative IOD years \u0026ndash; particularly 1998 and 2005 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed) \u0026ndash; a limitation also noted by Grunseich et al. (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Moreover, during extreme negative IOD event of 2016, the SDMI exhibits only a muted signal (green circles in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ed), whereas the SDI effectively captures the pronounced SSS anomalies associated with that event (green circles in Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb). These findings demonstrate that the SDI provides a more robust and dynamically consistent measures of IOD-related salinity variability than the conventional SDMI, highlighting the improved sensitivity of our newly defined index to both positive and negative IOD extremes. To further support the observational results, we have analyzed the CMIP6 models and found consistent results (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e and Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e3.2. How does the IOD produce dipole variability in SSS across the Indian Ocean?\u003c/h2\u003e \u003cp\u003eTo understand the physical processes underlying IOD-induced salinity variability, we examined how changes in ocean-atmosphere interactions during IOD events shape the SSS pattern (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003e). The regression results show that the IOD explains a substantial portion of the SSS variability in the tropical IO. During positive IOD events, low-SSS anomalies dominate the CEIO, while high-SSS anomalies occur in the SJC region (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eb and Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea). In the SJC region, the strengthening of southeasterly winds along the coast promotes coastal upwelling (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea), which shoals the thermocline and reduces the thickness of the barrier layer. This vertical restructuring facilitates the entrainment of subsurface, high-salinity water into the mixed layer (Sun et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The upwelled water, characterized by low temperature, high nutrients, and high salinity, contributes to surface cooling and enhances biological productivity. Consequently, the combined effects of upwelling and advection lead to increased salinity in the SJC during positive IOD events, while the opposite processes occur during negative IOD phases (Shi \u0026amp; Wang, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The regression of halocline depth supports this mechanism, showing pronounced shoaling in the SJC region during positive IOD conditions (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ed). Simultaneously, SST cooling in the SJC suppresses local convection and reduces precipitation, further contributing to salinity enhancement. Precipitations decrease markedly in the SJC during positive IOD (Fig.\u0026nbsp;\u003cspan refid=\"Fig7\" class=\"InternalRef\"\u003e7\u003c/span\u003ea), whereas precipitation increases over the CEIO region, contributing to surface freshening. However, consistent with previous findings (Yuhong et al., \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e2013\u003c/span\u003e), changes in precipitation-evaporation balance seem to exert only a weak influence on the SJC and CEIO SSS variability. Therefore, IOD-related SSS variability in tropical IO is likely to arise from a combination of processes other than precipitation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo further examine the physical processes involved in the SSS dipole, we examined the mixed-layer salt budget. Following Li et al. (\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2016\u003c/span\u003e), four key terms \u0026ndash; surface freshwater flux, zonal advection, meridional advection, and vertical entrainment \u0026ndash; were considered in the salt budget framework. Since entrainment contributes marginally compared to the other terms, we focus primarily on zonal advection, meridional advection, and surface freshwater flux as the dominant processes influencing the salt budget, while entrainment is included in the residual term to explain the overall SSS dipole mechanism. The SSS anomalies typically develop during autumn, and previous studies have shown that these anomalies lag mixed-layer salinity variations by about 1\u0026ndash;2 months (Li et al., \u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Accordingly, we analyzed the mixed-layer salt budget for the August\u0026ndash;September (AS, Fig. S5) and August\u0026ndash;October (ASO, Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e) periods.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e shows composites of the salt budget components averaged over five positive IOD (1994, 1997, 2006, 2015, and 2023) and six negative IOD (1996, 1998, 2005, 2010, 2016, and 2022) events during ASO. Among the horizontal advection terms, zonal advection dominates over meridional advection, indicating that zonal currents play a leading role in modulating SSS anomalies in the equatorial IO. During positive IOD events, the CEIO exhibits a negative SSS anomaly tendency primarily driven by anomalous zonal advection \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\left(-\\left.\\varvec{u}\\frac{\\:\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{x}}\\right)\\right.\\)\u003c/span\u003e\u003c/span\u003e, with a secondary contribution from anomalous meridional advection \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\left(\\left.-\\varvec{v}\\frac{\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{y}}\\right)\\right.\\)\u003c/span\u003e\u003c/span\u003e. In contrast, the SJC shows a positive SSS anomaly tendency resulting from anomalous meridional advection and the anomalous freshwater flux \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:\\left(-\\varvec{S}\\frac{\\left(\\varvec{P}-\\varvec{E}\\right)}{\\varvec{h}}\\right)\\)\u003c/span\u003e\u003c/span\u003e. During negative IOD events, a positive SSS anomaly tendency emerges in the CEIO, predominantly influenced by \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:-\\varvec{u}\\frac{\\:\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{x}}\\)\u003c/span\u003e\u003c/span\u003e, while \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:-\\varvec{v}\\frac{\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{y}}\\)\u003c/span\u003e\u003c/span\u003e acts in the opposite direction. Meanwhile the SJC exhibits a negative SSS anomaly tendency caused by \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:-\\varvec{v}\\frac{\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{y}}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:-\\varvec{S}\\frac{\\left(\\varvec{P}-\\varvec{E}\\right)}{\\varvec{h}}\\)\u003c/span\u003e\u003c/span\u003e. In both IOD phases, the SJC \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:-\\varvec{u}\\frac{\\:\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{x}}\\)\u003c/span\u003e\u003c/span\u003e term shows an opposite response relative to the observed SSS anomaly tendency. As shown in figure S5, the AS composites display a similar pattern: during positive IOD events, CEIO negative SSS anomaly tendency arises mainly from \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:-\\varvec{u}\\frac{\\:\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{x}}\\)\u003c/span\u003e\u003c/span\u003e with secondary contribution from \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:-\\varvec{v}\\frac{\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{y}}\\)\u003c/span\u003e\u003c/span\u003e, while the SJC positive SSS anomaly tendency is driven by \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:-\\varvec{v}\\frac{\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{y}}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:-\\varvec{S}\\frac{\\left(\\varvec{P}-\\varvec{E}\\right)}{\\varvec{h}}\\)\u003c/span\u003e\u003c/span\u003e. In contrast, during negative IOD events, positive SSS anomaly tendency in the CEIO is dominated by \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:-\\varvec{u}\\frac{\\:\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{x}}\\)\u003c/span\u003e\u003c/span\u003e and secondarily by \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:-\\varvec{v}\\frac{\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{y}}\\)\u003c/span\u003e\u003c/span\u003e, whereas the SJC negative tendency is driven by \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:-\\varvec{v}\\frac{\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{y}}\\)\u003c/span\u003e\u003c/span\u003e and \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:-\\varvec{S}\\frac{\\left(\\varvec{P}-\\varvec{E}\\right)}{\\varvec{h}}\\)\u003c/span\u003e\u003c/span\u003e. However, \u003cspan class=\"InlineEquation\"\u003e\u003cspan class=\"mathinline\"\u003e\\(\\:-\\varvec{u}\\frac{\\:\\partial\\:\\varvec{S}}{\\partial\\:\\varvec{x}}\\)\u003c/span\u003e\u003c/span\u003e acts in the opposite direction against the negative SSSA tendency. Notably, the amplitude of the negative IOD SSS anomaly response is weaker than that of the positive IOD (Fig.\u0026nbsp;\u003cspan refid=\"Fig8\" class=\"InternalRef\"\u003e8\u003c/span\u003e and Fig. S5).\u003c/p\u003e \u003c/div\u003e"},{"header":"4. Discussion and Conclusion","content":"\u003cp\u003eBased on long-term observational datasets, this study investigated IOD-induced changes in SSS in the tropical IO and introduced a refined Salinity Dipole Index (SDI) that better captures these changes than the previously proposed Salinity Dipole Mode Index (SDMI). Our results reveal a robust SSS dipole during boreal Autumn, characterized by a negative SSS anomaly in the CEIO and a positive anomaly in SJC during positive IOD years, and the opposite pattern during the negative IOD years. The concept of a salinity-based IOD (SIOD) proposed by Shi \u0026amp; Wang, (\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) extends the framework of the conventional SST-based DMI (Saji et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1999\u003c/span\u003e) and the biologically derived BDMI (Shi \u0026amp; Wang \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2021\u003c/span\u003e, \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). While BIOD and SIOD represent distinct aspects of the IOD, their regions differ from the conventional IOD domains and therefore require refined definitions to capture the respective IOD-related variability. For instance, the biological dipole and its related index (BDI) were defined between the south of the Indian Subcontinent and west of Sumatra (Pathirana et al., \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2024\u003c/span\u003ea, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2024\u003c/span\u003eb; Abeywickrama et al., \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). In this study, we propose a refined SDI based on CEIO\u0026ndash;SJC regions to more accurately capture salinity variations associated with the IOD.\u003c/p\u003e \u003cp\u003eAlthough a dipole pattern is evident in SSS responses to the IOD, the regions used in the SDMI (Shi and Wang, \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2025\u003c/span\u003e) may not accurately represent underlying salinity variability. The conventional IOD boxes (Saji et al., \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e1999\u003c/span\u003e) are optimized for SST variations and therefore inadequately capture salinity signals: the western box lies too far offshore to reflect the low-salinity waters of the Bay of Bengal (BoB), while the eastern box largely reflects upwelling-related SSS anomaly in the SJC. Our refined SDI effectively captures both positive and negative IOD events and shows a stronger correlation with the DMI (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.93) than the SDMI (r\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.82), particularly concerning negative IODs. Notably, while the SDMI failed to capture anomalies during the negative IOD events of 1998 and 2005, the SDI successfully reproduced them. This highlights that careful regional selection enhances the sensitivity and reliability of salinity-based indices. This improvement is consistent with previous findings by Grunseich et al. (\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2011\u003c/span\u003e). Furthermore, the SDMI also underrepresents salinity variability during extreme negative IOD events, such as in 2016, highlighting the advantage of the refined SDI.\u003c/p\u003e \u003cp\u003eAnalysis of the mixed-layer salt budget demonstrates that zonal advection is the dominant mechanism governing changes in salinity in the CEIO, whereas SJC variability is primarily controlled by meridional advection and freshwater flux. A non-linear relationship is observed between positive and negative IOD phases, with stronger salinity responses during positive IOD events, suggesting nonlinear salinity feedback in the IO. The dominance of zonal advection reflects the influence of strong equatorial currents and the dynamics of equatorial Kelvin and Rossby waves. In contrast, the SJC is more affected by wind-driven coastal upwelling, precipitation and barrier layer processes. These regional contrasts highlight the spatial heterogeneity of salinity drivers within the IOD system. Our findings also emphasize the asymmetric strength of IOD-induced SSS variations: positive IODs exhibit stronger air\u0026ndash;sea coupling and intensified easterly wind anomalies across the equator, whereas negative IODs are generally weaker and more susceptible to ENSO-induced changes in the Walker circulation (Cai \u0026amp; Cowan, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Sun et al., \u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e2022\u003c/span\u003e). The stronger amplitude of positive IOD-related salinity tendencies observed here is consistent with this asymmetric behavior. Furthermore, our salt budget analysis confirms that the magnitude of negative IOD SSS anomaly tendencies is smaller than during positive events, thus providing a dynamical explanation for the observed asymmetry.\u003c/p\u003e \u003cp\u003eOne striking feature is the contrasting SSS anomaly response observed along the eastern Indian coastline: enhanced salinity during positive IOD events and reduced salinity during negative IOD events. This pattern arises from large-scale anomalous currents modulated by zonal wind variations over the eastern IO. During negative (positive) IOD phases, anomalous westerlies (easterlies) generate downwelling (upwelling) Kelvin waves, strengthening (weakening) the Wyrtki jet and inducing basin-wide cyclonic (anticyclonic) circulation in the Bay of Bengal. These processes strengthen (weaken) the East India Coastal Current, driving the southward propagation of negative (positive) SSS anomalies along the eastern Indian coast (Chen et al., \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Together with the identified CEIO\u0026ndash;SJC dipole, this establishes a tripole-like SSS response across the IO during the IOD peak phase.\u003c/p\u003e \u003cp\u003eIn conclusion, this study provides a comprehensive understanding of salinity variability associated with the IOD and offers a refined SDI that captures both positive and negative events with greater accuracy. By linking surface salinity changes to underlying physical processes, we emphasize the important role of advection and freshwater flux in modulating the SSS dipole. The identification of asymmetric salinity responses and the emerging tripole structure highlights the complexity of air\u0026ndash;sea interactions in the tropical Indian Ocean. These findings improve the characterization of the IOD and provide valuable insights for the future modelling and prediction of ocean\u0026ndash;atmosphere coupled dynamics in a changing climate.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e \u003cp\u003eNot Applicable\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConsent for publication\u003c/strong\u003e \u003cp\u003eNot Applicable\u003c/p\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eConflict of interest\u003c/strong\u003e \u003cp\u003eThe authors declare that they have no conflicts of interest, financial or otherwise, related to the research, authorship, and/or publication of this article.\u003c/p\u003e \u003c/p\u003e\u003ch2\u003eAcknowledgement\u003c/h2\u003e \u003cp\u003eWe thank the World Climate Research Program and associated teams for providing the CMIP6 data. All graphs and analyses were performed using Python v. 3.12 and supportive packages.\u003c/p\u003e\u003ch2\u003eData Availability Statement\u003c/h2\u003e \u003cp\u003eThe CMIP6 data used in this study is available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://esgf-node.llnl.gov/projects/cmip6/\u003c/span\u003e\u003cspan address=\"https://esgf-node.llnl.gov/projects/cmip6/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. ECMWF data is available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.ecmwf.int/en/forecasts/datasets\u003c/span\u003e\u003cspan address=\"https://www.ecmwf.int/en/forecasts/datasets\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Salinity data is available at \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://data.marine.copernicus.eu/products\u003c/span\u003e\u003cspan address=\"https://data.marine.copernicus.eu/products\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAbeywickrama T, Pathirana G, Noh K-M, Dissanayake K, Wang D, Lee D-G (2025) Asymmetric effects of Indian ocean dipole on surface chlorophyll variability in the Indian ocean. Ocean Dyn 75(8):69. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s10236-025-01714-2\u003c/span\u003e\u003cspan address=\"10.1007/s10236-025-01714-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAshok K, Guan Z, Yamagata T (2001) Impact of the Indian Ocean dipole on the relationship between the Indian monsoon rainfall and ENSO. Geophys Res Lett 28(23):4499\u0026ndash;4502. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2001GL013294\u003c/span\u003e\u003cspan address=\"10.1029/2001GL013294\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBalaguru K, Foltz GR, Leung LR, Emanuel KA (2016) Global warming-induced upper-ocean freshening and the intensification of super typhoons. Nat Commun 7(1):13670. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/ncomms13670\u003c/span\u003e\u003cspan address=\"10.1038/ncomms13670\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBehera SK, Luo J-J, Masson S, Delecluse P, Gualdi S, Navarra A, Yamagata T (2005) Paramount Impact of the Indian Ocean Dipole on the East African Short Rains: A CGCM Study. J Clim 18(21):4514\u0026ndash;4530. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1175/JCLI3541.1\u003c/span\u003e\u003cspan address=\"10.1175/JCLI3541.1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBosc C, Delcroix T, Maes C (2009) Barrier layer variability in the western Pacific warm pool from 2000 to 2007. J Geophys Research: Oceans 114(C6). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2008JC005187\u003c/span\u003e\u003cspan address=\"10.1029/2008JC005187\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCai W, Cowan T (2013) Why is the amplitude of the Indian Ocean Dipole overly large in CMIP3 and CMIP5 climate models? Geophys Res Lett 40(6):1200\u0026ndash;1205. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/grl.50208\u003c/span\u003e\u003cspan address=\"10.1002/grl.50208\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen B, Qin H, Chen G, Xue H (2019) Ocean Salinity as a Precursor of Summer Rainfall over the East Asian Monsoon Region. J Clim 32(17):5659\u0026ndash;5676. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1175/JCLI-D-18-0756.1\u003c/span\u003e\u003cspan address=\"10.1175/JCLI-D-18-0756.1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen S, Qiu F, Jing C, Qiu Y, Zhang J (2024) Unraveling the Influence of Equatorial Waves on Post-Monsoon Sea Surface Salinity Anomalies in the Bay of Bengal. Remote Sens 16(8):1348. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/rs16081348\u003c/span\u003e\u003cspan address=\"10.3390/rs16081348\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCopernicus Marine Environment Monitoring Service (CMEMS) (2024) \u003cem\u003eGlobal Ocean Ensemble Physics Reanalysis (1993\u0026ndash;2023)\u003c/em\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ede Boyer Mont\u0026eacute;gut C, Madec G, Fischer AS, Lazar A, Iudicone D (2004) Mixed layer depth over the global ocean: An examination of profile data and a profile-based climatology. J Geophys Research: Oceans 109(C12). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2004JC002378\u003c/span\u003e\u003cspan address=\"10.1029/2004JC002378\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDelcroix T, Henin C, Porte V, Arkin P (1996) Precipitation and sea-surface salinity in the tropical Pacific Ocean. Deep Sea Res Part I 43(7):1123\u0026ndash;1141. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/0967-0637(96)00048-9\u003c/span\u003e\u003cspan address=\"10.1016/0967-0637(96)00048-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDoney SC, Ruckelshaus M, Emmett Duffy J, Barry JP, Chan F, English CA, Galindo HM, Grebmeier JM, Hollowed AB, Knowlton N, Polovina J, Rabalais NN, Sydeman WJ, Talley LD (2012) Climate Change Impacts on Marine Ecosystems. Annual Rev Mar Sci 4(1):11\u0026ndash;37. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1146/annurev-marine-041911-111611\u003c/span\u003e\u003cspan address=\"10.1146/annurev-marine-041911-111611\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDu Y, Zhang Y (2015) Satellite and Argo Observed Surface Salinity Variations in the Tropical Indian Ocean and Their Association with the Indian Ocean Dipole Mode. J Clim 28(2):695\u0026ndash;713. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1175/JCLI-D-14-00435.1\u003c/span\u003e\u003cspan address=\"10.1175/JCLI-D-14-00435.1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrunseich G, Subrahmanyam B, Murty VSN, Giese BS (2011) Sea surface salinity variability during the Indian Ocean Dipole and ENSO events in the tropical Indian Ocean. J Geophys Research: Oceans 116(C11). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2011JC007456\u003c/span\u003e\u003cspan address=\"10.1029/2011JC007456\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHersbach H (2023) B. B. ; B. P. ; B. G. ; H. A. ; M. S. J. ; N. J. ; P. C. ; R. R. ; R. I. ; S. D. ; S. A. ; S. C. ; D. D. ; T. J.-N. \u003cem\u003eERA5 monthly averaged data on single levels from 1940 to present\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.24381/cds.f17050d7\u003c/span\u003e\u003cspan address=\"10.24381/cds.f17050d7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHu S, Zhang Y, Feng M, Du Y, Sprintall J, Wang F, Hu D, Xie Q, Chai F (2019) Interannual to Decadal Variability of Upper-Ocean Salinity in the Southern Indian Ocean and the Role of the Indonesian Throughflow. J Clim 32(19):6403\u0026ndash;6421. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1175/JCLI-D-19-0056.1\u003c/span\u003e\u003cspan address=\"10.1175/JCLI-D-19-0056.1\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJensen TG (2001) Arabian Sea and Bay of Bengal exchange of salt and tracers in an ocean model. Geophys Res Lett 28(20):3967\u0026ndash;3970. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2001GL013422\u003c/span\u003e\u003cspan address=\"10.1029/2001GL013422\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJinadasa SUP, Pathirana G, Ranasinghe PN, Centurioni L, Hormann V (2020) Monsoonal impact on circulation pathways in the Indian Ocean. Acta Oceanol Sin 39(3):103\u0026ndash;112. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s13131-020-1557-5\u003c/span\u003e\u003cspan address=\"10.1007/s13131-020-1557-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKumar SP, Prasad TG (1999) Formation and spreading of Arabian Sea high-salinity water mass. J Geophys Research: Oceans 104(C1):1455\u0026ndash;1464. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/1998JC900022\u003c/span\u003e\u003cspan address=\"10.1029/1998JC900022\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLagerloef GSE (2002) Introduction to the special section: The role of surface salinity on upper ocean dynamics, air-sea interaction and climate. J Geophys Research: Oceans 107(C12). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2002JC001669\u003c/span\u003e\u003cspan address=\"10.1029/2002JC001669\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi J, Liang C, Tang Y, Dong C, Chen D, Liu X, Jin W (2016) A new dipole index of the salinity anomalies of the tropical Indian Ocean. Sci Rep 6(1):24260. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/srep24260\u003c/span\u003e\u003cspan address=\"10.1038/srep24260\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi Z, Lian T, Ying J, Zhu X, Papa F, Xie H, Long Y (2021) The Cause of an Extremely Low Salinity Anomaly in the Bay of Bengal During 2012 Spring. J Geophys Research: Oceans 126(10). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2021JC017361\u003c/span\u003e\u003cspan address=\"10.1029/2021JC017361\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePathirana G, Min Noh K, Lee D-G, Park H-J, Wang D, Kug J-S (2024) Intensified surface chlorophyll responses to the Indian Ocean Dipole under greenhouse warming. Environ Res Lett 19(6):064072. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1088/1748-9326/ad4db8\u003c/span\u003e\u003cspan address=\"10.1088/1748-9326/ad4db8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePathirana G, Noh KM, Lee D-G, Lee H, Kug J-S (2024) A biological dipole variability in the Indian Ocean. Environ Res Lett 19(1):014070. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1088/1748-9326/ad1641\u003c/span\u003e\u003cspan address=\"10.1088/1748-9326/ad1641\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePerigaud C, McCreary JP, Zhang KQ (2003) Impact of interannual rainfall anomalies on Indian Ocean salinity and temperature variability. J Geophys Research: Oceans 108(C10). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2002JC001699\u003c/span\u003e\u003cspan address=\"10.1029/2002JC001699\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRao RR, Kumar KVS (1991) Evolution of salinity field in the upper layers of the east central Arabian Sea and northern Bay of Bengal during summer monsoon experiments. J Earth Syst Sci 100(1):69\u0026ndash;78. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/BF02843485\u003c/span\u003e\u003cspan address=\"10.1007/BF02843485\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRao RR, Sivakumar R (2003) Seasonal variability of sea surface salinity and salt budget of the mixed layer of the north Indian Ocean. J Geophys Research: Oceans 108(C1). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2001JC000907\u003c/span\u003e\u003cspan address=\"10.1029/2001JC000907\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaji NH, Goswami BN, Vinayachandran PN, Yamagata T (1999) A dipole mode in the tropical Indian Ocean. Nature 401(6751):360\u0026ndash;363. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/43854\u003c/span\u003e\u003cspan address=\"10.1038/43854\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSaji N, Yamagata T (2003) Possible impacts of Indian Ocean Dipole mode events on global climate. Climate Res 25:151\u0026ndash;169. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3354/cr025151\u003c/span\u003e\u003cspan address=\"10.3354/cr025151\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSandeep KK, Pant V, Girishkumar MS, Rao AD (2018) Impact of riverine freshwater forcing on the sea surface salinity simulations in the Indian Ocean. J Mar Syst 185:40\u0026ndash;58. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jmarsys.2018.05.002\u003c/span\u003e\u003cspan address=\"10.1016/j.jmarsys.2018.05.002\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchott FA, McCreary JP (2001) The monsoon circulation of the Indian Ocean. Prog Oceanogr 51(1):1\u0026ndash;123. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0079-6611(01)00083-0\u003c/span\u003e\u003cspan address=\"10.1016/S0079-6611(01)00083-0\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShi W, Wang M (2021) A biological Indian Ocean Dipole event in 2019. Sci Rep 11(1):2452. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41598-021-81410-5\u003c/span\u003e\u003cspan address=\"10.1038/s41598-021-81410-5\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShi W, Wang M (2022) Biological dipole mode indices: New parameters to characterize the physical and biological processes of the Indian Ocean Dipole event. Prog Oceanogr 206:102847. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.pocean.2022.102847\u003c/span\u003e\u003cspan address=\"10.1016/j.pocean.2022.102847\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShi W, Wang M (2025) Salinity Indian Ocean Dipole: Another facet of the Indian Ocean Dipole phenomenon from satellite remote sensing. Sci Remote Sens 11:100184. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.srs.2024.100184\u003c/span\u003e\u003cspan address=\"10.1016/j.srs.2024.100184\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStammer D, Martins MS, K\u0026ouml;hler J, K\u0026ouml;hl A (2021) How well do we know ocean salinity and its changes? Prog Oceanogr 190:102478. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.pocean.2020.102478\u003c/span\u003e\u003cspan address=\"10.1016/j.pocean.2020.102478\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSubrahmanyam B, Murty VSN, Heffner DM (2011) Sea surface salinity variability in the tropical Indian Ocean. Remote Sens Environ 115(3):944\u0026ndash;956. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.rse.2010.12.004\u003c/span\u003e\u003cspan address=\"10.1016/j.rse.2010.12.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSun Q, Zhang Y, Du Y, Jiang X (2022) Asymmetric Response of Sea Surface Salinity to Extreme Positive and Negative Indian Ocean Dipole in the Southern Tropical Indian Ocean. J Geophys Research: Oceans 127(11). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1029/2022JC018986\u003c/span\u003e\u003cspan address=\"10.1029/2022JC018986\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThompson B, Gnanaseelan C, Salvekar PS (2006) Variability in the Indian Ocean circulation and salinity and its impact on SST anomalies during dipole events. J Mar Res 64(6):853\u0026ndash;880. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1357/002224006779698350\u003c/span\u003e\u003cspan address=\"10.1357/002224006779698350\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWyrtki K (1973) An Equatorial Jet in the Indian Ocean. Science 181(4096):262\u0026ndash;264. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1126/science.181.4096.262\u003c/span\u003e\u003cspan address=\"10.1126/science.181.4096.262\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYuan C, Yamagata T (2015) Impacts of IOD, ENSO and ENSO Modoki on the Australian Winter Wheat Yields in Recent Decades. Sci Rep 5(1):17252. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/srep17252\u003c/span\u003e\u003cspan address=\"10.1038/srep17252\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYuhong Z, Yan D, Shaojun Z, Yali Y, Xuhua C (2013) Impact of Indian Ocean Dipole on the salinity budget in the equatorial Indian Ocean. J Geophys Research: Oceans 118(10):4911\u0026ndash;4923. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1002/jgrc.20392\u003c/span\u003e\u003cspan address=\"10.1002/jgrc.20392\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang Y, Du Y, Qu T (2016) A sea surface salinity dipole mode in the tropical Indian Ocean. Clim Dyn 47(7\u0026ndash;8):2573\u0026ndash;2585. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s00382-016-2984-z\u003c/span\u003e\u003cspan address=\"10.1007/s00382-016-2984-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"climate-dynamics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"cldy","sideBox":"Learn more about [Climate Dynamics](https://www.springer.com/journal/382)","snPcode":"382","submissionUrl":"https://submission.nature.com/new-submission/382/3","title":"Climate Dynamics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false},"keywords":"Sea Surface Salinity, Indian Ocean Dipole, salinity dipole index","lastPublishedDoi":"10.21203/rs.3.rs-8401709/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8401709/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eThe Indian Ocean Dipole (IOD) is a dominant coupled ocean\u0026ndash;atmosphere mode in the tropical Indian Ocean, characterized by opposing sea surface temperature anomalies (SSTAs) between the western and southeastern regions. In addition to its well-established climatic and biological impacts, the IOD strongly modulates sea surface salinity (SSS), a key variable governing ocean circulation and air\u0026ndash;sea interactions. However, the nature of IOD-induced salinity changes and the appropriate formulation of a salinity dipole index (SDI) remain debated, particularly across different IOD phases. Previous studies have proposed contrasting definitions of salinity dipole regions: one, based on high-resolution eddy-resolving model simulations, identifies the dipole between the Central Equatorial Indian Ocean (CEIO: 70\u0026deg;E\u0026ndash;90\u0026deg;E, 5\u0026deg;S\u0026ndash;5\u0026deg;N) and the Sumatra\u0026ndash;Java Coast (SJC: 100\u0026deg;E\u0026ndash;110\u0026deg;E, 13\u0026deg;S\u0026ndash;3\u0026deg;S); while another, derived from satellite-based SSS observations, aligns with the conventional IOD regions. Here, we revisit the IOD\u0026ndash;SSS relationship and reassess the SDI using long-term observational data and historical simulations from a Coupled Model Intercomparison Project Phase 6 (CMIP6) model datasets. Our results show that the IOD-induced salinity dipole emerges most clearly during boreal autumn but is not co-located with the conventional IOD regions. Instead, it consistently develops between the CEIO and SJC, particularly during strong positive IOD events, exhibiting low salinity in the CEIO and high salinity along the SJC. This dipole structure is primarily controlled by IOD-driven surface water advection and freshwater flux anomalies. Based on these findings, we propose a refined definition of the salinity dipole regions and SDI, which provides a more robust and consistent representation of salinity variability during IOD events.\u003c/p\u003e","manuscriptTitle":"A Refined Perspective on Indian Ocean Dipole–Driven Surface Salinity Changes and the Salinity Dipole Index","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-04 09:07:25","doi":"10.21203/rs.3.rs-8401709/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"reviewerAgreed","content":"","date":"2026-02-03T08:33:08+00:00","index":0,"fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-02-02T22:00:21+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-12-22T05:18:13+00:00","index":"","fulltext":""},{"type":"submitted","content":"Climate Dynamics","date":"2025-12-19T02:09:30+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"climate-dynamics","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"cldy","sideBox":"Learn more about [Climate Dynamics](https://www.springer.com/journal/382)","snPcode":"382","submissionUrl":"https://submission.nature.com/new-submission/382/3","title":"Climate Dynamics","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"Springer Hybrid","inReviewEnabled":true,"inReviewRevisionsEnabled":false}}],"origin":"","ownerIdentity":"43225fc8-5663-4a48-85c9-109e7cca06ce","owner":[],"postedDate":"February 4th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-02-04T09:07:25+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-04 09:07:25","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8401709","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8401709","identity":"rs-8401709","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}