Abstract
The Wyrtki Jet plays an important role in the tropical Indian Ocean circulation, yet its variability near the eastern boundary was observed scarcely. In this study, a strong Wyrtki Jet event in the spring 2014 is analyzed using satellite-tracked surface drifter trajectories combined with the mooring data in the eastern Indian Ocean. The observed maximum zonal velocity reaches 1.2 m s-1, twice as much as the spring jet climatology. Following the strong spring jet, another eastward jet with velocity of 0.9 m s-1 is observed in July, when the seasonal zonal currents typically flow westward. The LICOM ocean general circulation model, which successfully simulates the observed zonal currents in the eastern Indian Ocean in 2014, is used to investigate the dynamics of the Wyrtki Jet event. Decomposed equatorial waves from the model simulation show that the 2014 event is dominated by wind-driven downwelling Kelvin waves and the reflected upwelling Rossby waves of earlier wind forcing in the central and eastern basin. Reconstructed zonal currents from the wave coefficients suggest that the second baroclinic-mode Rossby waves reflected from the Kelvin waves dominate the strong spring Wyrtki Jet, with higher-order baroclinic modes contributing more than 30% of the velocity amplitude. The eastward current in July 2014 is dominated by an intraseasonal wind-driven Kelvin wave. The disclosed dynamics highlight the important roles of the boundary reflections and higher-order modes in modulating the ocean circulation in the eastern Indian Ocean near the boundary.
1 Introduction
The Wyrtki Jet is a strong eastward surface current in the tropical Indian Ocean appearing during monsoon transitions in boreal spring and fall, identified first by . With strong eastward currents, the Wyrtki Jet transports warm water from central Indian Ocean to the eastern basin, directly influencing heat redistribution and ocean circulation in the eastern Indian Ocean warm pool, and furthermore impacts the global climate variability by modulating atmospheric convections (; ; ; ).
Early study of the Wyrtki Jet is based primarily on moored current observations (; ; ; ), showing that the Wyrtki Jet in fall is generally stronger than in spring. The maximum velocity of the jets is observed at 80.5°E south of Siri Lanka in fall 1993 exceeding 1.5 m s-1, while the spring jet is much weaker with velocity of 0.5 m s-1 (; ). In the eastern Indian Ocean, the surface and subsurface Wyrtki Jets have been studied using the ADCP data from the Triangle Trans-Ocean Buoy Network (TRITON) and Research Moored Array for African-Asian-Australian Monsoon Analysis and Prediction (RAMA, ) programs (; ; ; ; ; ; ; ; , ; ).
In addition to mooring observations, satellite-tracked surface drifter trajectories have also been used to study tropical circulation in the Indian Ocean. However, most drifters were deployed in the western Indian Ocean (; ; ). Since the 21st century, the Argos floats and Surface Velocity Program drifters have been widely used to investigate surface circulation in the central Indian Ocean, with limited attention paid to the Wyrtki Jet close to the Indian Ocean eastern boundary (; ; ; ; ).
Previous studies have pointed out that the semiannual zonal currents in the equatorial Indian Ocean, which is the most significant characteristic of the Wyrtki Jet, is forced by the wind-driven equatorial Kelvin and Rossby waves (; ; ; ; ; ; ). The semiannual wind is found to resonate with the propagation of the second baroclinic-mode equatorial Kelvin and Rossby waves in the central basin (; , ; ). Early studies have focused on the role of the eastern boundary reflection in the study of the semiannual resonance (). have shown that the reflection of the semiannual Rossby waves into the equatorial Kelvin waves at the Indian Ocean western boundary are important for the resonant response. These studies show that the reflected waves enhance the wind-forced semiannual zonal currents at the central basin, whereas they largely cancel each other near the eastern boundary. The ship-drift data have shown that the Wyrtki Jets are weak in the eastern Indian Ocean east of 80°E (), consistent with the above dynamics.
In this study, a strong Wyrtki Jet event in the eastern Indian Ocean near the eastern boundary in the spring of 2014 is observed using satellite-tracked surface drifter trajectories and the RAMA mooring data, in contrast to the ship-drift climatology. An ocean general circulation model is used to simulate the observed currents, based on which equatorial wave decomposition is applied to the model simulation to investigate the underlying wave dynamics. Data and model used in this study and the wave decomposition method are introduced in Section 2. The analysis of the surface currents and the associated wave dynamics are presented in Section 3. Discussions are given in Section 4 followed by summary in Section 5.
2 Data and methods
2.1 Data
2.1.1 Observational data
Surface ocean currents derived from drifter observations are used to analyze the Wyrtki Jet in this study. From April 8 to May 15, 2014, eighteen satellite-tracked drifters were deployed in the eastern Indian Ocean along two sections: the equatorial section from 82°E to 92°E, and the eastern boundary section from (6°S, 96°24′E) to (4°N, 90°44′E) along the western continental slope of Sumatra (Figure 1). The surface drifters manufactured by the Chinese Xiaolong Instruments, LLC, contain a module of Global Positioning System (GPS) and satellite communication in a floating ball and a tube-shaped drogue at 15 m below the sea surface. The configuration of the drifters follows the World Ocean Circulation Experiment (WOCE) standard, with a ratio of 82 between the surface area of the drogue and that of the semi-sphere ball exposed in the air. The design allows the drifters to follow the ocean surface currents instead of the winds. Two different position-fixing frequencies are used for the floats, transmitting signals back to the ground through the Globalstar satellites at 1-hour and 8-hour intervals, respectively. The accuracy of the GPS positioning is about 100 m in the open oceans.
Figure 1
The climatological ship-drift current data, Mariano Global Surface Velocity Analysis 1.0 (MGSVA) produced by , are used to compare with the Lagrangian current vectors calculated from the drifter trajectories. The ship-drift data are monthly means with a horizontal resolution of 1°×1°, covering the global oceans.
The current data observed from RAMA moorings are also used to compare with the drifter observations and validate the model simulation. Velocities at 40 m depth collected from moorings located at 80.5°E and 90°E on the equator are used.
2.1.2 Ocean model
The LASG/IAP Climate system Ocean model (LICOM) is used to simulate the ocean currents in the eastern Indian Ocean following . The model is developed by the State Key Laboratory of Numerical Modeling for Atmospheric Sciences and Geophysical Fluid Dynamics, Institute of Atmospheric Physics, Chinese Academy of Sciences (LASG/IAP). The horizontal and vertical resolution of the model is 0.5° and 30 layers from 12.5 m to 5319.08 m. The model was spun up for 900 years using the World Ocean Atlas 1994 climatological field, and forced by the 6-hour wind and surface flux from National Climatic Data Center (NCDC) from January 2001 to December 2015 as a hindcast run. Monthly output of the 3-D velocity field and sea level pressure are used to investigate the dynamics of the surface currents. To further investigate the Wyrtki Jet in the spring of 2014, daily output of surface currents is also used for the year 2014.
2.2 Method: wave decomposition
The wave decomposition method is used in this study to understand roles of equatorial Kelvin and Rossby waves in the variability of surface currents. Following , the 3-D velocity and dynamic height are projected onto the eigenfunctions to extract each baroclinic mode:
Where
is the eigenfunction which represents the vertical structure of the ocean.
and
for each baroclinic mode can be further decomposed into meridional modes to get the coefficients of equatorial Kelvin and Rossby waves ():
where is the meridional structure of the Kelvin and Rossby waves in terms of th order Hermite function, and is the wave coefficients. Based on the wave decomposition, the Kelvin or Rossby wave equation is then integrated with wind forcings from the decomposed or reflected wave coefficients at the boundary to investigate the contributions of wind forcing or boundary reflection to the current variability.
3 Results
3.1 Surface currents in the eastern Indian Ocean: drifter observations
Eighteen satellite-tracked drifters drogued at 15 m were deployed from April 8 to May 15 to capture surface circulation in the eastern Indian Ocean. Among them, six drifters were deployed along the eastern boundary and the rest twelve along the equator (Figure 1). Most of them continued working for more than two months, provided Lagrangian measurements of the surface currents in the boundary and equatorial region during the whole spring season.
The trajectories of the drifters show that all six drifters deployed along the equator rapidly moved eastward toward the eastern boundary, reflecting the influence of the Wyrtki Jet. Upon reaching the eastern boundary (~96.5°E), the drifters bifurcated with three of them moved southward and three of them moved northward along the boundary, suggesting branching of the Wyrtki Jet at the eastern boundary near 96.5°E on the equator. Among those drifters flowed to the south, one ran aground at the coast of Sumatra on April 29. Another drifter joined the South Java Current (SJC), traveled southeastward along the coast of Sumatra to ~10°S, and finally turned westward carried by the South Equatorial Current (SEC). The third drifter traveled southwestward from the eastern boundary.
Of the three northward-moving drifters, one died one month after deployment. Another one made a small circle to the north and then moved southwestward across the equator. The third one moved northward to ~2°N, then turned back westward along 2°N carried by a countercurrent. It subsequently rejoined the Wyrtki Jet at 84°E and propagated eastward again, eventually crossing the equator near 96°E into the southern hemisphere. Overall, all the six drifters, except the two dead, eventually ended up in the southern hemisphere and moved westward in July, suggesting cross-equatorial transport at the eastern boundary. Furthermore, the observed trajectories clearly capture the eastward propagation of equatorial Kelvin waves and their reflections at the eastern boundary into westward-propagating Rossby waves. The results also indicate that most Kelvin waves reaching the eastern boundary transform into coastal Kelvin waves propagating southeastward along the Sumatra-Java coast, while only a small portion propagates northward. This is consistent with satellite altimetry observations reported by ; , and .
For the twelve drifters deployed along the east boundary, all of them traveled eastward to the coast then moved northward or southward, indicating the bifurcation of the Wyrtki Jet at the eastern boundary. Notably, all six drifters deployed in the northern hemisphere eventually ran aground along the coast. This is because the northward boundary currents along Sumatra are relatively weak, leading to the grounding of all drifters. In contrast, in the southern hemisphere, the drifters traveled southeastward carried by the strong SJC, then turn westward and entered the open ocean under the influence of westward-propagated Rossby waves. The stronger southward boundary currents allow the drifters deployed in the south to work for a longer duration than those deployed in the north.
The Lagrangian velocities are calculated based on these trajectories and compared with historical MGSVA ship-drift data during the same season (Figure 2). The results show that a strong Wyrtki Jet is captured by the drifters between 85°E and 90°E along the equator, with a maximum zonal velocity of ~1.2 m s-¹, substantially exceeding the climatological mean (~0.5 m s-¹) derived from ship-drift data. Specifically, the jet maintains strong velocity of ~0.8 m s-¹ in the boundary region east of 90°E, and only weakens east of 95°E at the boundary. Although strong Wyrtki Jets with velocities of ~1.2 m s-¹ have been reported before (e.g., ; ), they were typically observed in the central Indian Ocean near ~80°E, or in the fall season when the Wyrtki Jet is generally stronger than the spring jet. None of them are in the eastern Indian Ocean, especially in regions close to the boundary.
Figure 2
The Lagrangian velocity field also clearly illustrates the bifurcation of the Wyrtki Jet at the eastern boundary and its contribution to the westward SEC in the southern Indian Ocean (Figure 2a). In contrast, due to the coarse spatial and temporal resolution, detailed structures of the boundary currents cannot be resolved well in the ship-drift climatology (Figure 2b). This highlights the advantage and necessity of surface drifters in observing boundary currents and coastal circulations.
3.2 Surface currents in the eastern Indian Ocean: model simulation and validation
To find out the dynamics of the strong Wyrtki Jet in the eastern Indian Ocean, equatorial waves of different modes are decomposed from a LICOM model simulation. In this section, surface currents simulated by LICOM model are validated by comparing them with climatological ship-drift data, drifter observations, and RAMA mooring data.
Figure 3 shows the comparison of the climatological zonal velocity from ship-drift data (a) and simulated by LICOM from 2001 to 2015 (b). The seasonal cycle of the zonal velocity is well reproduced by LICOM simulation, characterized by eastward jets in spring and fall during monsoon transitions, and westward currents in summer and winter. The Wyrtki Jet in spring occurs from March to May, with a maximum velocity exceeding 0.6 m s-¹ in early May, consistent with the ship-drift data. In fall after September, the circulation is dominated by the fall Wyrtki Jet, with a maximum velocity of ~0.5 m s-¹, weaker than the ship-drift observation (~0.7 m s-¹). The maximum velocities of the Wyrtki Jets are located near 80°E in both datasets.
Figure 3
In summer and winter, the model simulation shows weak westward currents (~0.2 m s-¹, Figure 3b), while such reversed zonal flows are only evident in the central basin west of 80°E in the ship-drift data (figure omitted). This difference could be due to the complex ocean processes in the boundary region, where both observation and model simulation have larger uncertainties.
In addition to the climatological velocity, we also examine the daily output of the model in the year 2014, as shown in Figure 3c. The spring Wyrtki Jet in 2014 exhibits a maximum eastward velocity exceeding 1.2 m s-¹, approximately twice as large as the climatological mean in Figure 3b. Similar to the climatology, the maximum occurs in May and around 80°E. However, in the area east of 80°E, the jet persists strong eastward currents of ~1.0 m s-¹ extend to 90°E, instead of weakening significantly like the climatology. This indicates that the model successfully reproduces the intensified Wyrtki Jet in the eastern Indian Ocean in the spring of 2014. Notably, another eastward current with a velocity of 0.6 m s-¹ appears in July, when the climatological eastward current vanishes or reverses. Dynamics of this eastward current will be further discussed below.
To further evaluate the model simulation of the strong spring Wyrtki Jet in 2014, we identified the days in April and May when drifter observations were available, then averaged the model daily output over the same days. The mean velocities are compared with the Lagrangian velocities derived from drifter trajectories in April and May, respectively (Figure 4). The comparison shows good agreement between the two datasets with a strong eastward jet extending to the eastern boundary (~95°E). The model slightly underestimates the eastward velocity, and the northwestward boundary current along the northern tip of the Sumatra coast, likely due to the temporal averaging and the relatively coarse spatial resolution of the model (0.5°). Other surface circulation patterns in the eastern Indian Ocean, including the bifurcation of the Wyrtki Jet at the eastern boundary, the southeastward boundary current along the Sumatra-Java coasts (SJC), and the westward SEC between 10° and 15°S, are all reproduced well by the model. Overall, the surface currents in the eastern Indian Ocean in the spring of 2014 are successfully simulated by the LICOM ocean model as compared with drifter observations.
Figure 4
The RAMA mooring array is critical in the tropical Indian Ocean observation system since it provides sustained time series of ocean and climate variables across the entire basin. In the eastern equatorial Indian Ocean, current measurements from RAMA moorings are available at (0°, 80.5°E) and (0°, 90°E). Here we first compare the monthly model output with RAMA data through 2001 to 2015 (Figures 5a, b). The results show a good agreement between the two datasets. The LICOM model successfully reproduces the amplitude of the Wyrtki Jets in most of years, although several strong events with zonal velocities exceeding 1 m s-¹ are underestimated (e.g., fall jets in 2005, 2010, and 2014). The spring jets, however, are generally well simulated by the model in amplitude and variability.
Figure 5
We also compare the daily model output in 2014 with RAMA data (Figures 5c, d). At 80.5°E on the equator, the model agrees well with observations, with a correlation coefficient of 0.85 (Figure 5c). The evolution of the Wyrtki Jet from April to June is well reproduced, including two velocity peaks exceeding 1.4-1.5 m s-¹ in early and mid-May. However, the model slightly underestimates the fall jet (October-December) and the zonal currents in March.
At 90°E, RAMA mooring data for 2014 are largely unavailable. Instead, we use Lagrangian velocities estimated from drifter trajectories passing by this location to compare with the model output (Figure 5d). Both the model and drifter data show eastward currents of ~1.2 m s-¹ in late April to early May, as well as another eastward flow in July, except the model velocity (~0.6 m s-¹) of the July jet is weaker than drifter estimates (~0.9 m s-¹).
3.3 The wave dynamics in the model simulation
The LICOM simulated velocity and dynamic height are decomposed into equatorial Kelvin and Rossby waves to investigate the dynamics of the strong surface jet. In this section, wave dynamics including reflection at the eastern boundary and wind forcings will be analyzed and discussed.
3.3.1 Decomposed wave coefficients
Fifty baroclinic modes in total are decomposed from the LICOM model output. Here we take the first baroclinic mode as an example. The second baroclinic mode shares similar structure of the first baroclinic mode but with larger amplitudes due to basin resonance (). Higher-order modes are not discussed in detail here, as their amplitudes are relatively small compared with the first two baroclinic modes. Figure 6 shows the Kelvin and the first meridional-mode Rossby wave coefficients of the first baroclinic mode decomposed from the LICOM output in 2014. Positive and negative coefficients represent downwelling and upwelling waves, respectively. The Kelvin wave coefficients show significant semiannual variation in the eastern Indian Ocean including the boundary region (Figure 6a). The downwelling Kelvin waves dominate the eastern Indian Ocean basin from April to mid-July and from October to December, while summer and winter seasons are dominated by the upwelling Kelvin waves. In particular, strong downwelling Kelvin waves propagate eastward reaching the boundary area in May, corresponding to the strong Wyrtki Jet event observed by the drifters in spring in this region. Another strong intraseasonal downwelling Kelvin wave event is captured as well in early July, may be related to the eastward jet during this period. The propagation speed agrees well with the theoretical phase speed of the first baroclinic mode Kelvin wave, as shown in the black solid line at the bottom of Figure 6a. According to the wind stress projected onto the Kelvin wave of the first baroclinic mode (Figure 6b), these Kelvin waves are mainly generated and strengthened by the wind forcing, especially the intraseasonal Kelvin wave in July.
Figure 6
The first meridional-mode Rossby wave coefficients of the first baroclinic mode also exhibit a pronounced semiannual oscillation (Figure 6c), and in phase with the Kelvin waves reaching the eastern boundary. This indicates that downwelling/upwelling Rossby waves are generated by the reflection of downwelling/upwelling Kelvin waves at the eastern boundary. After leaving the boundary, the first meridional Rossby waves propagate westward at a speed approximately one-third of the Kelvin wave phase speed (black solid line at the bottom of the figure), consistent with linear wave theory. In spring of 2014, upwelling Rossby wave reflected from the upwelling Kelvin waves in winter is enhanced in the eastern basin between 85°E and 90°E (Figure 6c) due to anomalous westerlies (Figure 6d), which significantly contributes to the strong Wyrtki Jet together with the downwelling Kelvin wave.
3.3.2 Roles of boundary reflection and wind forcing
To quantify the different contributions of the reflection at the eastern boundary and wind forcing to the Rossby wave, we integrate the Rossby wave equation (, where represents the wave coefficients, is the Rossby wave speed, and represents the wind stress) along the characteristic line 1) with the wind stress projected onto the first meridional-mode Rossby wave start from Rossby wave coefficients at the eastern boundary (Figure 7b), and 2) with the same wind forcing but start from zero (Figure 7d), which represents the contribution of the wind forcing only. Differences between 1) and 2) can be regarded as the contribution of the boundary reflection (Figure 7c). Here we use 96°E as the idealized meridional boundary following , since the eastern boundary of the Indian Ocean is not a meridional boundary. The results indicate that the seasonal oscillation and strength of the Rossby wave in boundary region are primarily controlled by the reflection of the Kelvin waves at the eastern boundary. In particular, the phase reversal of the Rossby wave coefficients between May and June suggests that eastern boundary reflection plays an important role in modulating the direction reversal of the Wyrtki Jet during the monsoon season. The wind forcing, however, is only important in the open ocean west of 90°E, where it contributes equally to the amplitudes of the Rossby waves with the boundary reflection.
Figure 7
Figure 8 shows the comparison of the first meridional-mode Rossby wave coefficient derived from linear reflection of the Kelvin wave (black), and directly decomposed from the model output (green). According to the linear wave theory, the reflected first meridional-mode Rossby waves are 1.22 of the Kelvin wave amplitudes (). Thus, the Kelvin wave coefficient at the eastern boundary is multiplied by 1.22 to obtain the linear reflected Rossby wave coefficient. The comparison shows that the decomposed first meridional-mode Rossby wave agrees well with the linear reflection of the Kelvin wave at the eastern boundary in spring. In contrast, amplitude of the decomposed Rossby wave in July is only half of that from linear reflection. The zonal current during this period thus remains dominated by Kelvin waves instead of Rossby waves and flows eastward.
Figure 8
3.4 Reconstructed surface zonal currents in the eastern Indian Ocean
Based on the decomposed wave coefficients from the LICOM model output, the zonal currents from April to July 2014 were reconstructed to investigate the individual contribution of different baroclinic modes to the surface currents. Since higher-order meridional modes contribute very little, only the Kelvin wave and the first meridional-mode Rossby wave of the first ten baroclinic modes are used in the reconstruction (Figure 9).
Figure 9
The sum of the first ten baroclinic modes successfully reproduces the zonal currents simulated by the model, with small differences in amplitude (Figure 9b). In the full model output shown in Figure 9a, the maximum velocity of the spring Wyrtki Jet exceeds 1.2 m s-¹ at 80°E, whereas the reconstructed currents from the first ten baroclinic modes have a maximum of approximately 1.1 m s-¹ (Figure 9b). In addition, the weak westward flow in June and the reappearance of eastward flow in July are both reconstructed, although the July eastward current is weaker than that in the full model output. This indicates that the first ten baroclinic modes largely determine the phase and variability of surface currents in the eastern Indian Ocean, while higher-order modes contribute to the amplitudes.
To further examine contributions of individual baroclinic modes, zonal velocities reconstructed from each baroclinic modes are shown in Figures 9c–f. The results show that the first and second baroclinic modes exhibit similar zonal current variations, with eastward currents in April and May, and westward currents in June and July (Figures 9c, d), in phase with the monsoon transition. This suggests that the lower-order modes are primarily controlled by wind-driven linear processes, consistent with the wave dynamics analyzed in the previous sections. The amplitude of the currents in the second mode is much larger than that in the first mode, indicating that the strength of the zonal currents is mostly controlled by the second baroclinic mode. The first baroclinic mode contributes about 16% of the total eastward jet, with a maximum velocity of about 0.2 m s-¹. The second baroclinic mode provides the dominant contribution, accounting for nearly 50% of the total velocity exceeding 0.5 m s-¹. Thus, the lower-order modes determine the primary seasonal variation of the spring Wyrtki Jet, while higher-order modes play an important role in modulating its amplitude, contributing more than 30% of the total velocity.
The dominant role of the second baroclinic mode is consistent with the basin resonance theory (, ; ). The theoretical phase speed of the second baroclinic mode in the wave decomposition in this study is 1.74 m s-1, corresponding to a basin mode period of ~160 days, close to the unique semiannual period of the wind forcing over the equatorial Indian Ocean. In comparison, the corresponding basin mode period for the first baroclinic mode is ~102 days and therefore less resonated with the semiannual wind forcing.
The third to tenth baroclinic modes show a significantly different structure with eastward flow persists throughout the whole period (Figures 9e, f). The eastward current in July is clearly shown in the reconstruction from these higher-order modes, suggesting the dominance of these modes in the July jet. Besides, the westward propagation signals are evident in the first four baroclinic modes and less apparent in modes higher than four, consistent with the previous studies (e.g., ).
Different roles of the equatorial Kelvin waves and the first meridional-mode Rossby waves are also investigated by reconstructing the zonal currents from April to July 2014 using the Kelvin and first meridional-mode Rossby wave coefficients, respectively, as shown in Figures 10, 11. The results indicate that the spring Wyrtki Jet in the eastern Indian Ocean is dominated by the Rossby waves, which contributes ~0.6 m s-¹ of the total velocity, and the second baroclinic mode contributes the most (Figure 11). Kelvin waves contribute approximately 40% of the spring jet (Figure 10). The weaker jet observed in early July, however, is mainly associated with Kelvin waves (Figure 10). As discussed above, anomalous westerly winds generated intraseasonal downwelling Kelvin waves not fully reflected into the downwelling Rossby waves (Figure 8). As a result, the zonal current remained dominated by Kelvin waves and continued flowing eastward, rather than reversing into the westward current.
Figure 10
Figure 11
4 Discussions
4.1 Intraseasonal wind stress anomaly in 2014
The anomalous eastward jet in July 2014 is found to be dominated by a strong intraseasonal downwelling Kelvin wave forced by intraseasonal westerly winds. In this section, intraseasonal wind stress anomalies in 2014 are investigated and possible connection with background climate variability are discussed.
Figure 12 shows the daily zonal wind stress anomaly and its intraseasonal oscillation (30–90-day band-pass filter) in 2014, together with the Madden-Julian Oscillation (MJO, Madden and Julian, 1971) index. The anomalies are based on the 2001–2015 climatology. In the eastern Indian Ocean east of 80°E, strong westerly wind anomalies are found in summer from mid-June to July, and at the end of December (Figure 12b), consistent with the intraseasonal Kelvin wave events (Figure 6a). These anomalies appear to propagate eastward from the western and central Indian Ocean, suggesting a possible connection with the MJO, as it is the dominant mode of intraseasonal variability in the tropical atmosphere characterized by eastward propagation. Figure 12a shows the evolution of the MJO in 2014 based on the real time multivariate Madden Julian index (RMM). In summer, an active MJO event developed in June and propagated eastward, with its maximum amplitude over the Maritime Continent in late June. The enhanced convection is suggested to induce the westerly wind anomalies in the eastern Indian Ocean later in July associated with the intraseasonal downwelling Kelvin wave. Similarly, another MJO event is found in December, corresponding to the eastward-propagated westerly wind anomaly during this period.
Figure 12
4.2 Interannual variability of the Wyrtki Jet
Large-scale climate modes in tropics, such as the El Niño-Southern Oscillation (ENSO) and the Indian Ocean Dipole (IOD), can influence the strength of the Wyrtki Jet by modulating the background climate conditions over the equatorial Indian Ocean. To examine the potential roles of these climate modes, we investigate the interannual variability of the Wyrtki Jet and zonal wind stress in the eastern Indian Ocean during 2001 to 2015 (Figure 13). In general, the zonal wind stress anomaly in the eastern Indian Ocean (80°-95°E) exhibits opposite variations with the dipole mode index (DMI), with westerly/easterly wind stress anomaly during negative/positive IOD events. For example, the easterly wind anomaly appears during the 2006 positive IOD event, and extremely strong westerly wind stress anomaly is observed during the 2010 negative IOD event. The zonal currents anomaly in the eastern Indian Ocean, however, shows relatively weak correlations with either IOD or ENSO indices. As shown in Figure 13a, both ENSO and IOD are in neutral phases in 2014, suggesting the anomalous Wyrtki Jet events in spring and summer 2014 may not directly controlled by ENSO or IOD variability.
Figure 13
It is worth mentioning that an early negative IOD event appears to have developed in 2014, rather than a canonical IOD event. Previous studies have pointed out that this new type of IOD develops in May-June and matures in July-August, and is only observed since the mid-1970s forced by the rapid Indian Ocean warming (). In 2014, the negative summer IOD may contribute to the anomalous westerly wind stress during early summer (Figure 13b), which in turn lead to the eastward anomalies in zonal currents from spring to July in 2014 in the eastern Indian Ocean (Figure 13c).
4.3 Comparison with the 2013 spring Wyrtki Jet
Since the 2010s, the spring Wyrtki Jet in the eastern Indian Ocean in 2013 is also found to be a strong Wyrtki Jet event as reported by . The unsmoothed eastward current anomaly at 90°E on the equator in spring 2013 is found larger than that in 2014 in the LICOM simulation (Figure 13C). However, the 2013 event differs significantly from the 2014 event, as the eastward Wyrtki Jet anomaly only appears in May, whereas the 2014 event exhibits a strong jet both in the spring and in July. As a result, a persistent strong eastward anomalous jet from April to July is observed in the 2014 event (Figure 13c). The corresponding wind fields are also different: The peaks of the westerly wind stress anomalies are observed in May in 2013, and in July in 2014 (Figure 13b), which drive the anomalously strong Wyrtki Jet in May 2013 and July 2014, respectively. However, a strong Wyrtki Jet is observed in the spring of 2014, under the weak westerly wind stress anomaly forcing, suggesting different forcing from the eastern boundary reflection in 2014 with that in 2013.
Different forcings of boundary reflection in the strong Wyrtki Jet events of 2013 and 2014 are shown in the interannual Kelvin and first meridional-mode Rossby wave coefficients (Figure 14). In 2013, the eastern Indian Ocean is first dominated by a strong downwelling Kelvin wave, and then an upwelling Kelvin wave. While in 2014, an upwelling Kelvin wave appears first during January-March, before the strong downwelling Kelvin wave in spring. Consequently, the upwelling Rossby waves reflected from the upwelling Kelvin wave at the eastern boundary in the spring of 2014 superposed on the downwelling Kelvin waves to force the strong eastward Wyrtki Jet near the boundary. In contrast, during the same period in 2013, no upwelling signal is observed and reflected so that the Indian Ocean basin is dominated by the downwelling Kelvin waves and reflected downwelling Rossby waves. The reflected Rossby waves with the same sea level phase but opposite zonal equatorial currents cancel the Kelvin wave zonal currents, resulting in a weaker Wyrtki Jet near the boundary. These results indicate that the anomalously strong Wyrtki Jet event in 2013 is dominated by the wind-driven downwelling Kelvin wave and its reflection, while the 2014 event is dominated by wind-driven downwelling Kelvin waves and the reflected upwelling Rossby waves of earlier wind forcing in the central and eastern basin, which contribute ~0.6 m s-1 of the total velocity as shown in the reconstructed zonal currents in Section 3.
Figure 14
5 Summary
A strong Wyrtki Jet in the spring of 2014 comparable to that in the central basin in fall was observed in the eastern Indian Ocean using satellite-tracked surface drifter observations. The maximum zonal velocity of the Wyrtki Jet estimated from the drifter trajectories reaches 1.2 m s-¹ between 85°E and 90°E, significantly exceeding the climatological mean derived from MGSVA ship-drift data. Following the strong spring jet, another eastward jet was observed in July of 2014 during the monsoon season, when the climatology usually flows to the west.
The wave dynamics of the eastward jets are investigated using the LICOM model output. The surface current simulated by LICOM model compared well with the ship-drift climatology, in-situ observations in 2014 from drifters, and RAMA moorings. The model successfully reproduced the strong Wyrtki Jet in spring and the eastward jet in July in the eastern Indian Ocean in 2014, except that the amplitude of the eastward velocity is smaller than the observations.
The wave coefficients of the first baroclinic mode decomposed from the LICOM model output show strong downwelling Kelvin waves and upwelling Rossby wave in the eastern boundary region in the spring of 2014, associated with the strong Wyrtki Jet event. In July, an intraseasonal downwelling Kelvin wave is excited by anomalous westerly wind. The downwelling and upwelling Kelvin waves were reflected into the Rossby waves at the eastern boundary in agreement with the linear dynamics, except for the intraseasonal Kelvin waves in July. Analyses of the wind forcing and boundary reflection suggest that the boundary reflection determines the direction and amplitudes of the zonal currents near the eastern boundary, while local wind forcing plays a secondary role.
Reconstruction of the zonal currents using the decomposed wave coefficients reveals that the first two baroclinic modes dominate the seasonal variability of the Wyrtki Jet. Higher-order modes are found to contribute more than 30% of the velocity amplitude, twice that of the first baroclinic mode (16%). The strong spring jet is primarily driven by the second baroclinic-mode equatorial upwelling Rossby waves reflected from the equatorial upwelling Kelvin waves. The eastward current in July is mainly attributed to a strong intraseasonal downwelling Kelvin wave forced by intraseasonal westerly winds. The westerly wind anomaly is associated with the MJO activity and an early negative IOD event took place in summer. Only half of the Kelvin waves are reflected into Rossby waves during the July event based on the decomposed wave coefficients of both the first and second baroclinic mode. It is worth mentioning that the decomposed coefficients could have underestimated the intraseasonal Rossby waves due to the equatorial geostrophic assumption in the decomposition method. Detailed investigations of boundary reflection for high-frequency waves and their contributions to intraseasonal variability of the Wyrtki Jet are postponed to a future study.
Statements
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Author contributions
XH: Writing – review & editing, Methodology, Formal analysis, Investigation, Writing – original draft. DY: Resources, Project administration, Funding acquisition, Data curation, Conceptualization, Supervision, Writing – review & editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the National Key Research and Development Program of China (2019YFC1509102, 2020YFA0608800), the National Natural Science Foundation of China (NSFC, 42561144236, 42550074), and the Ocean Negative Carbon Emission (ONCE) program. The cruise is supported by the NSFC Shiptime Sharing Project (NORC2014-09).
Acknowledgments
We thank the crews of R/V SHIYAN 1 and scientists from South China Sea Institute of Oceanology, Chinese Academy of Sciences, especially Dr. Jian Li, for the successful cruise and deployment of the drifters.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
1
BealL.HormannV.LumpkinR.FoltzG. (2013). The response of the surface circulation of the Arabian Sea to monsoonal forcing. J. Phys. Oceanogr.43, 2008–2022. doi: 10.1175/JPO-D-13-033.1
2
BoulangerJ. P.MenkesC. (1995). Propagation and reflection of long equatorial waves in the Pacific Ocean during the 1992-1993 El Niño. J. Geophys. Research.: Oceans.100, 25041–25059. doi: 10.1029/95JC02956
3
CaoG.XuT.WeiZ. (2024). Seasonal differences of Wyrtki Jet intraseasonal variabilities. Front. Mar. Sci.11, 1517779. doi: 10.3389/fmars.2024.1517779
4
ChenG.HanW.HuangR. X.WangD. (2024). Equatorial waves substantially modulate currents in the tropical Indian Ocean. Innovation Geosci.2, 100053. doi: 10.59717/j.xinn-geo.2024.100053
5
ChenG.HanW.LiY.WangD.McPhadenM. J. (2015). Seasonal-to-interannual time scale dynamics of the equatorial undercurrent in the Indian Ocean. J. Phys. Oceanogr.45, 1532–1552. doi: 10.1175/JPO-D-14-0225.1
6
DuY.CaiW.WuY. (2013). A new type of the Indian Ocean dipole since the mid-1970s. J. Clim.26, 959–972. doi: 10.1175/JCLI-D-12-00047.1
7
DuanY.LiuL.HanG.LiuH.YuW.YangG.et al. (2016). Anomalous behaviors of Wyrtki Jets in the equatorial Indian Ocean during 2013. Sci. Rep.6, 29688. doi: 10.1038/srep29688
8
GrodskyS. A.CartonJ. A.MurtuguddeR. (2001). Anomalous surface currents in the tropical Indian Ocean. Geophys. Res. Lett.28, 4207–4210. doi: 10.1029/2001GL013592
9
HanW.McCrearyJ. P.AndersonD. L. T.MarianoA. J. (1999). Dynamics of the eastern surface jets in the equatorial Indian Ocean. J. Phys. Oceanogr.29, 2191–2209. doi: 10.1175/1520-0485(1999)029<2191:DOTESJ>2.0.CO;2
10
HanW.McCrearyJ. P.MasumotoY.VialardJ.DuncanB. (2011). Basin resonances in the equatorial Indian Ocean. J. Phys. Oceanogr.41, 1252–1270. doi: 10.1175/2011JPO4591.1
11
HaseH.MasumotoY.KurodaY.MizunoK. (2008). Semiannual variability in temperature and salinity observed by Triangle Trans-Ocean Buoy Network (TRITON) buoys in the eastern tropical Indian Ocean. J. Geophys. Research.: Oceans.113, C01016. doi: 10.1029/2006JC004026
12
HuangK.HanW.WangD.WangW.XieQ.ChenJ.et al. (2018). Features of the equatorial intermediate current associated with basin resonance in the Indian Ocean. J. Phys. Oceanogr.48, 1333–1347. doi: 10.1175/JPO-D-17-0238.1
13
HuangK.WangD.HanW.FengM.ChenG.WangW.et al. (2019). Semiannual variability of middepth zonal currents along 5°N in the eastern Indian Ocean: characteristics and causes. J. Phys. Oceanogr.49, 2715–2729. doi: 10.1175/JPO-D-19-0089.1
14
HuangK.WangD.QiuY.WuY.LiuK.HuangB.et al. (2023). Meridional-width variability of near-equatorial zonal currents along 80.5°E on seasonal to interannual timescales in the Indian Ocean. J. Geophys. Research.: Oceans.128, e2022JC019147. doi: 10.1029/2022JC019147
15
IskandarI.MasumotoY.MizunoK. (2009). Subsurface equatorial zonal current in the eastern Indian Ocean. J. Geophys. Research.: Oceans.114, C06005. doi: 10.1029/2008JC005188
16
IskandarI.McphadenM. J. (2011). Dynamics of wind-forced intraseasonal zonal current variations in the equatorial Indian Ocean. J. Geophys. Research.: Oceans.116, C06019. doi: 10.1029/2010JC006864
17
JensenT. G. (1993). Equatorial variability and resonance in a wind-driven Indian Ocean model. J. Geophys. Research.: Oceans.98, 22533–22552. doi: 10.1029/93JC02565
18
KnoxR. A. (1976). On a long series of measurements of Indian Ocean equatorial currents near Addu Atoll. Deep-Sea. Res.23, 211–221. doi: 10.1016/0011-7471(76)91325-5
19
LuytenJ. R.FieuxM.GonellaJ. (1980). Equatorial currents in the western Indian Ocean. Science209, 600–603. doi: 10.1126/science.209.4456.600
20
LuytenJ. R.RoemmichD. H. (1982). Equatorial Currents at semi-annual period in the Indian Ocean. J. Phys. Oceanogr.12, 406–413. doi: 10.1175/1520-0485(1982)012<0406:ecasap>2.0.co;2
21
MaddenR. A.JulianP. R. (1971). Detection of a 40–50 day oscillation in the zonal wind in the tropical Pacific. J. Atmospheric Sci.28(5), 702-708. doi: 10.1175/1520-0469(1971)028<0702:DOADOI>2.0.CO;2
22
MarianoA. J.RyanE. H.PerkinsB. D.SmithersS. (1995). “ The Mariano global surface velocity analysis 1.0,” in USCG Report No. CG-D-34-95, Washington, D.C.
23
MasumotoY.HaseH.KurodaY.MatsuuraH.TakeuchiK. (2005). Intraseasonal variability in the upper layer currents observed in the eastern equatorial Indian Ocean. Geophys. Res. Lett.32, L02607. doi: 10.1029/2004GL021896
24
McCrearyJ. P. (1981). A linear stratified ocean model of the equatorial undercurrent. Philos. Trans. R. Soc. A. Mathematical. Phys. Eng. Sci.298, 603–635. doi: 10.1098/rsta.1981.0002
25
McPhadenM. (1982). Variability in the central equatorial Indian Ocean. I. Ocean dynamics. J. Mar. Res.40, 157–176.
26
McPhadenM. J.MeyersG.AndoK.MasumotoY.MurtyV.RavichandranM.et al. (2009). RAMA: the research moored array for African–Asian–Australian monsoon analysis and prediction. Bull. Am. Meteorol. Soc90, 459–480. doi: 10.1175/2008BAMS2608.1
27
MolinariR. L.OlsonD.ReverdinG. (1990). Surface current distributions in the tropical Indian Ocean derived from compilations of surface buoy trajectories. J. Geophys. Research.: Oceans.95, 7217–7238. doi: 10.1029/JC095iC05p07217
28
NaguraM.McPhadenM. J. (2010). Wyrtki Jet dynamics: Seasonal variability. J. Geophys. Research.: Ocean.115, C07009. doi: 10.1029/2009jc005922
29
NienhausM. J.SubrahmanyamB.MurtyV. (2012). Altimetric observations and model simulations of coastal Kelvin Waves in the Bay of Bengal. Mar. Geod.35, 190–216. doi: 10.1080/01490419.2012.718607
30
NyadjroE. S.McPhadenM. J. (2014). Variability of zonal currents in the eastern equatorial Indian Ocean on seasonal to interannual time scales. J. Geophys. Research.: Oceans.119, 7969–7986. doi: 10.1002/2014JC010380
31
PengS.QianY.-K.LumpkinR.DuY.WangD.LiP. (2015). Characteristics of the near-surface currents in the Indian Ocean as deduced from satellite-tracked surface drifters. Part I: Pseudo-Eulerian statistics. J. Phys. Oceanogr.45, 441–458. doi: 10.1175/JPO-D-14-0050.1
32
PrernaS.ChatterjeeA.MukherjeeA.RavichandranM.ShenoiS. (2019). Wyrtki Jets: Role of intraseasonal forcing. J. Earth Syst. Sci.128, 21. doi: 10.1007/s12040-018-1042-0
33
QiuY.LiL.YuW. (2009). Behavior of the Wyrtki Jet observed with surface drifting buoys and satellite altimeter. Geophys. Res. Lett.36, L18607. doi: 10.1029/2009GL039120
34
RaoR. R.Girish KumarM. S.RavichandranM.RaoA. R.GopalakrishnaV. V.ThadathilP. (2010). Interannual variability of Kelvin wave propagation in the wave guides of the equatorial Indian Ocean, the coastal Bay of Bengal and the southeastern Arabian Sea during 1993–2006. Deep. Sea. Res. Part. I.: Oceanographic. Res. Papers.57, 1–13. doi: 10.1016/j.dsr.2009.10.008
35
ReppinJ.SchottF. A.FischerJ.QuadfaselD. (1999). Equatorial currents and transports in the upper central Indian Ocean: Annual cycle and interannual variability. J. Geophys. Research.: Oceans.104, 15495–15514. doi: 10.1029/1999JC900093
36
ReverdinG. (1987). The upper equatorial Indian Ocean: The climatological seasonal cycle. J. Phys. Oceanogr.17, 903–927. doi: 10.1175/1520-0485(1987)017<0903:TUEIOT>2.0.CO;2
37
ReverdinG.FieuxM.GonellaJ.LuytenJ. (1983). “ Free drifting buoy measurements in the Indian Ocean equatorial jet,” in Elsevier Oceanography Series, (Amsterdam - Oxford - New York: Elsevier) vol. 36. , 99–120. doi: 10.1016/S0422-9894(08)70630-5
38
SchottF.ReppinJ.FischerJ.QuadfaselD. (1994). Currents and transports of the Monsoon Current south of Sri Lanka. J. Geophys. Research.: Oceans.99, 25127–25141. doi: 10.1029/94JC02216
39
SchottF. A.McCrearyJ. P. (2001). The monsoon circulation of the Indian Ocean. Prog. Oceanogr.51, 1–123. doi: 10.1016/S0079-6611(01)00083-0
40
ShenoiS.SajiP.AlmeidaA. (1999). Near-surface circulation and kinetic energy in the tropical Indian Ocean derived from Lagrangian drifters. J. Mar. Res.57, 885–907. doi: 10.1357/002224099321514088
41
WangJ.YuanD. (2015). Roles of western and eastern boundary reflections in the interannual sea level variations during negative Indian Ocean dipole events. J. Phys. Oceanogr.45, 1804–1821. doi: 10.1175/JPO-D-14-0124.1
42
WyrtkiK. (1973). An equatorial jet in the Indian Ocean. Science181, 262–264. doi: 10.1126/science.181.4096.262
43
XuT.WeiZ.CaoG.LiS. (2016). Pathways of intraseasonal Kelvin waves in the Indonesian throughflow regions derived from satellite altimeter observation. Atmos. Oceanic. Sci. Lett.9, 375–380. doi: 10.1080/16742834.2016.1208047
44
YuanD.HanW. (2006). Roles of equatorial waves and western boundary reflection in the seasonal circulation of the equatorial Indian Ocean. J. Phys. Oceanogr.36, 930–944. doi: 10.1175/JPO2905.1
45
YuanD.LiuH. (2009). Long wave dynamics of sea level variations during Indian Ocean dipole events. J. Phys. Oceanogr.39, 1115–1132. doi: 10.1175/2008JPO3900.1
46
ZhengS.-J.ZhangY.-H.ZhuangW.LiJ.-X.DuY. (2012). Typical surface seasonal circulation in the Indian Ocean derived from Argos floats. Atmos. Oceanic. Sci. Lett.5, 329–333. doi: 10.1080/16742834.2012.11447015
Summary
Keywords
boundary reflection, Eastern Indian Ocean, equatorial wave dynamics, Kelvin wave, Rossby wave, Wyrtki Jet
Citation
Hu X and Yuan D (2026) Dynamics of the strong Wyrtki Jet events in the eastern Indian Ocean in 2014. Front. Mar. Sci. 13:1874797. doi: 10.3389/fmars.2026.1874797
Received
07 May 2026
Revised
20 August 2026
Accepted
24 August 2026
Published
04 September 2026
Volume
13 - 2026
Edited by
Andrew Poje, The City University of New York, United States
Reviewed by
Shoichiro Kido, The University of Tokyo, Japan
Takaya Uchida, Moscow Institute of Physics and Technology, Russia
Updates
Copyright
© 2026 Hu and Yuan.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Dongliang Yuan, dyuan@tongji.edu.cn
Disclaimer
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.