Wang Etal JGR2022

21699291, 2022, 11, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.
1029/2022JC018694 by 1ST INSTITUTION OF OCEANOGRAPHY STATE OCEANIC ADMIN, Wiley Online Library on [01/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
RESEARCH ARTICLE Moored Observations of the Timor Passage Currents in the
10.1029/2022JC018694
Indonesian Seas
Special Section: Jing Wang1,2 , Zhengbei Zhang1,2 , Xiang Li3, Zheng Wang1 , Yao Li1, Jiajia Hao1,
Years of the Maritime Continent Xia Zhao1 , Corry Corvianawatie2,4 , Dewi Surinati4 , Dongliang Yuan1,2,5,6 , and Tengfei Xu5
Key Points:
1
CAS Key Laboratory of Ocean Circulation and Waves, Center for Ocean Mega-Science, Institute of Oceanology, Chinese
• T imor Passage currents, measured by Academy of Science, Function Laboratory for Ocean Dynamics and Climate, Pilot National Laboratory for Marine Science
a subsurface mooring, show strong and Technology (Qingdao), Qingdao, China, 2University of Chinese Academy of Sciences, Beijing, China, 3Southern Marine
annual and semiannual harmonics Science and Engineering Guangdong Laboratory (Zhuhai), Zhuhai, China, 4Research Center for Oceanography-National
in the upper and lower layers,
Research and Innovation Agency (BRIN), Jakarta, Indonesia, 5Key Laboratory of Marine Science and Numerical Modeling,
respectively
• Upward phase propagation of the First Institute of Oceanography, Ministry of Natural Resources, Function Laboratory for Regional Oceanography and
currents below 200 m is observed, Numerical Modeling, Pilot National Laboratory for Marine Science and Technology (Qingdao), Qingdao, China, 6Shandong
suggesting downward energy Key Laboratory of Marine Science and Numerical Modeling, Qingdao, China
propagation
• The total transport through the Timor
Passage features a stronger semiannual
cycle and a weak annual cycle
Abstract The structure and variability of the currents in the Timor Passage are studied based on the
measurements from a subsurface mooring deployed in the center of the passage from September 2017 to
December 2019. The annual mean currents are found to flow westward into the Indian Ocean in the upper
Supporting Information:
1,400 m, with a maximum velocity core at ∼50 m. The mean volume transport through the Timor Passage
Supporting Information may be found in
the online version of this article.
into the Indian Ocean is estimated to be −9.9 ± 1.0 Sv (1 Sv = 10 6 m 3 s −1) in the upper 1,400 m with much
of this transport concentrated in the upper 480 m (−8.9 Sv). This value contrasts with the estimate of −7.5 Sv
above 1,890 m based on INSTANT measurements. The transport during the 2019 positive Indian Ocean
Correspondence to:
D. Yuan,
Dipole event was only slightly larger than that during the 2018 normal year in the upper 480 m. The variability
dyuan@fio.org.cn in the along-strait velocity in the Timor Passage is dominated by an annual cycle in the upper 150 m and a
semiannual cycle in the lower layer. The annual cycle in the upper layer is mainly driven by local monsoonal
Citation: forcing. Downward energy propagation is observed below 200 m, which is shown to be mainly driven by
Wang, J., Zhang, Z., Li, X., Wang, Z., remote Kelvin waves from the Indian Ocean. The transport variability through the Timor Passage is dominated
Li, Y., Hao, J., et al. (2022). Moored by a semiannual cycle, associated with the semiannual circulation over the equatorial Indian Ocean, with the
observations of the Timor Passage
currents in the Indonesian seas. Journal
transport in the upper and lower layers largely canceling each other in the annual cycle.
of Geophysical Research: Oceans,
127, e2022JC018694. https://doi. Plain Language Summary Based on the INSTANT mooring measurements, the transport
org/10.1029/2022JC018694 through the Timor Passage contributes a large portion of the transport of the Indonesian Throughflow (ITF).
However, due to the scarcity of the observations, the transport and variability of the Timor currents have not
Received 30 MAR 2022
Accepted 3 NOV 2022 been adequately studied. In this study, we show that the currents in the upper 150 m, measured by an acoustic
Doppler current profiler moored at the valley center of the Timor Passage, feature strong annual harmonics
Author Contributions:
associated with local monsoonal forcing. Semiannual harmonics below 200 m reveal upward phase propagation
Conceptualization: Jing Wang associated with remote forcing from the Indian Ocean. The transport variability through the Timor Passage
Data curation: Xiang Li, Zheng Wang, features a stronger semiannual cycle in the upper 480 m instead of an annual cycle, as observed in the local
Yao Li, Corry Corvianawatie, Dewi annual winds, due to the mutual cancelation of the annual transports in the upper and lower layers. Thus,
Surinati
Formal analysis: Jing Wang the baroclinic processes play an important role in producing the semiannual-dominant variability of the ITF
Funding acquisition: Jing Wang, through the Timor Passage. The disclosed results are important for understanding the dynamics of the ITF
Tengfei Xu connection between the Indian Ocean and the Indonesian seas.
Investigation: Xiang Li, Corry
Corvianawatie, Dewi Surinati
Methodology: Jing Wang, Zhengbei
Zhang, Xiang Li 1. Introduction
Supervision: Jing Wang, Tengfei Xu
Visualization: Jing Wang, Zhengbei The Indonesian Throughflow (ITF) is the only conduit in the low latitudes connecting two ocean basins, the
Zhang tropical Indian Ocean and Pacific Ocean, and transports warm and fresh waters from the western Pacific Ocean
Writing – original draft: Jing Wang
into the eastern Indian Ocean through the Indonesian seas (Gordon et al., 2010; Hautala et al., 2001; Molcard
et al., 2001; Sprintall et al., 2009). There are three outflow straits through which water exits the Indonesian seas
into the Indian Ocean: the Lombok and Ombai Straits, and Timor Passage (Figure 1). The Timor Passage is a
© 2022. American Geophysical Union. long, narrow trench with a sill at 1,890-m depth and a width of 100 km between the Roti Island and the Australian
All Rights Reserved. continental shelf, leading to deep current transport into the Indian Ocean (Van Riehl, 1943).
WANG ET AL. 1 of 17
21699291, 2022, 11, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2022JC018694 by 1ST INSTITUTION OF OCEANOGRAPHY STATE OCEANIC ADMIN, Wiley Online Library on [01/12/2022]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
Journal of Geophysical Research: Oceans 10.1029/2022JC018694
Writing – review & editing: Xiang The currents in the Timor Passage have been observed before through mooring deployments, Shallow Pressure
Li, Zheng Wang, Jiajia Hao, Xia Zhao,
Gauge Array (SPGA), and shipboard acoustic Doppler current profile (sADCP) mapping. The moored current
Corry Corvianawatie, Dongliang Yuan,
Tengfei Xu meters show relatively steady, surface-intensified flows toward the Indian Ocean (Molcard et al., 1994, 1996).
The annual mean transport of the Timor Passage in the upper 200 m has been calculated to be −3.6 Sv based on
SPGA and sADCP data (Hautala et al., 2001). The transport in the Timor Passage has a large seasonal variation
due to the influence of the monsoon (Clarke & Liu, 1993; Masumoto & Yamagata, 1996). Long current time
series were observed from 2004 to 2006 during the INSTANT program. Four moorings from two deployments
were anchored from north to south in the Timor Passage at the Roti, sill, south slope, and Ashmore sites at depths
of 741, 1,890, 1,386, and 902 m, respectively, during December 2003 to June 2005 and June 2005 to December
2016. The seasonal cycle and interannual variability of the transport through the Timor Passage were discussed
briefly based on the INSTANT measurements (Sprintall et al., 2009). The mooring array across the eastern side
of Timor between April 2011 and December 2015 from the Australian Integrated Marine Observatory System
(IMOS) and CSIRO Oceans and Atmosphere is also used to estimate the Timor throughflow contribution to the
ITF and its seasonal cycle (Peña-Molino et al., 2022). The interannual variability in transport is closely related
to the El Niño-Southern Oscillation (ENSO), which is weak during El Niño events and strong during La Niña
events (Sprintall & Révelard, 2014). The ENSO and Indian Ocean Dipole (IOD) events coincided in 2006 during
the INSTANT measurements. The extreme positive IOD event in 2019 occurred without an ENSO event during
our mooring measurements. Recent study also suggests the important role of the Indian Ocean subsurface initial-
ization during 2019 IOD on the forecast of the 2020/2021 La Niña event (Wang et al., 2022). Therefore, the
influence of the ENSO and IOD on the transport in the upper 500 m through the Timor Passage can be quantified.
Numerical model results demonstrate that the transport of the outflow side occurs primarily in two vertical modes
in the upper 400 m (Potemra et al., 2003). The first mode is characterized by surface intensification that decays
downward, and the second mode is characterized by opposing flows in the upper 100 m and the underlying layers.
Recently, a mooring array of the Western Pacific Ocean Circulation-ITF (WPOC-ITF) constructed by the Insti-
tute of Oceanology, Chinese Academy of Sciences, was maintained in the Timor Passage during 2017–2019,
covering the strongest positive IOD event of 2019 on record (DOI et al., 2020), and the resulting data set serves
as an important complement to the INSTANT observations. In this study, the mean structure and variability of
the currents and transport in the Timor Passage are studied based on subsurface mooring measurements recorded
from September 2017 to December 2019. Section 2 provides a description of the mooring data and the analysis
methods employed in this study. Section 3 displays the mean and the variability of the currents and transport in
the Timor Passage. A discussion and the conclusions are presented in Sections 4 and 5, respectively.
2. Data and Methods

2.1. Data
2.1.1. Mooring Observations
The data analyzed in this paper were derived from the subsurface mooring located at 122.88°E, 11.26°S in the
Timor Passage (Figure 1). The mooring system includes three Aanderaa current meters at nominal depths of 750,
1,000, and 1,400 m; an upward-looking 75-kHz acoustic Doppler current profiler (ADCP) at a nominal depth
of 500 m; and five SBE37 CTDs at nominal depths of 250, 600, 750, 1,000, and 1,400 m. The mooring design
is shown in Figure 2. The ADCP data cover September 2017 to December 2019 with 60-bin measurements at
8-m intervals. The missing values were linearly interpolated in the vertical direction. The velocity data were then
linearly interpolated at a standard interval of 1 m. The hourly data were subsampled to daily data after Thompson
filtering (Thompson, 1983). The Aanderaa current meter data cover the period from 21 September 2017 to 10
May 2019, which is shorter than the time span of the ADCP data due to battery failure of the current meters. In
this paper, the data above 1,400 m were used for analysis.
The data of the mooring deployed at the Timor Sill (122.87°E, 11.27°S) during the INSTANT program are
compared with the mooring data obtained in this study. The top part of the Timor Sill mooring from the first
deployment prematurely separated in August 2004, but deeper current meters were later recovered during the
turn-around cruise in June 2005. Therefore, the Timor Sill mooring data include data from only four current
meters at depths of 539, 687, 987, and 1,784 m, and only a small amount of ADCP data. The Timor Sill mooring
WANG ET AL. 2 of 17
,7)
(TXDWRULDO
.:
&R :
DV
.
WD
OW
UD
SHS
G
,7)
)ORUHV6HD $ORU
)ORUHV
RU
7LP
6X
P
ED
L
RW
5
1:$XVWUDOLDQ6KHOI
Figure 1. Bathymetry in the Indonesian seas based on ETOPO1 data (upper). The Nusa Tenggara region, including the
Timor Passage, is labeled in a box. (lower) The zoomed-in box shows the mooring (triangle) deployed in the Timor Passage.
The 1,000-m and 2,000-m isobaths are also marked (contours). Equatorial and coastal-trapped Kelvin waves in the eastern
Indian Ocean are marked, and schematics of the Indonesian Throughflow (ITF) in the Western Pacific Ocean and Indonesian
seas are also shown with arrows.
data from the INSTANT cover the period from 30 December 2003 to 12 December 2006. The details of the moor-
ing design were outlined in Sprintall et al. (2009), and the data are more fully described in Cowley et al. (2008).
2.1.2. Other Data Sets
Delayed time (DT) daily sea level data derived from AVISO (Ducet et al., 2000) were obtained from the Coper-
nicus Marine Environment Monitoring Service (CMEMS; ftp.sltac.cls.r/Core/). The daily data cover the period
from 1 January 1993 to the present. Daily scatterometer wind stress data recorded by the advanced scatterometer
(ASCAT) in the Timor Passage from 2008 to 2019 were also used to calculate the climatological wind stress. The
merged gridded wind product of several satellite missions is available globally with a spatial resolution of 0.25° in
longitude and latitude (Bentamy & Fillon, 2012). The monthly wind stress data used herein for the model forcing
were derived from the fifth-generation reanalysis data of the European Center for Medium Range Weather Fore-
casts (ECMWF reanalysis V5, ERA5) from 1979 to 2020, with a spatial resolution of 0.25° longitude × 0.25° lati-
tude. The monthly Bluelink Reanalysis (BRAN2020) data set is used to cross-validate the interannual variability
in transport through the Timor Passage and ocean wave pathways in the study region. The reanalysis is a multi-
year integration of the Ocean Forecasting Australian Model during 1993–2021 with a horizontal resolution of
0.1° in longitude and latitude. The model has good performance in ITF studies from intraseasonal-to-interannual
variability and has been validated with the INSTANT data in several studies (Schiller et al., 2007, 2010).
WANG ET AL. 3 of 17
Figure 2. The subsurface mooring configuration in the Timor Passage (a) and data coverage during the mooring period (b).
2.2. Ocean Models

2.2.1. Linear Continuously Stratified Model
A simple, linear continuously stratified model (LCSM) was used to study the wave dynamics of current and
transport in the Timor Passage in this study. The model was the same as that used by Li et al. (2020) and closely
follows the theory described by McCreary (1981). The LCSM covers the tropical Indo-Pacific Ocean, includ-
ing the Indonesian seas between 30°S and 30°N with a horizontal resolution of 0.1° longitude × 0.1° latitude.
Three sensitivity runs with different forcing regions were conducted as summarized in Table 1: (a) IORUN1: the
model was forced west of 130°E and zero forcing was conducted in the tropical Pacific Ocean (east of 130°E);
(b) IORUN2: the model was forced west of 110°E and zero forcing was conducted in the tropical Pacific Ocean
(east of 110°E); and (c) INDOPACRUN: the LCSM was forced in the tropical Indo-Pacific Ocean (30°E−60°W,
30°S–30°N). Since the model is linear, the model simulation with only forcing in the Pacific Ocean (named
Table 1
Names and Descriptions of All the Pacemaker Experiments Used in This Study
Experiments
Names Descriptions
IORUN1 The model was forced west of 130°E and zero forcing was conducted in the
tropical Pacific Ocean (east of 130°E)
IORUN2 The model was forced west of 110°E and zero forcing was conducted in the
tropical Pacific Ocean (east of 110°E)
INDOPACRUN The model was forced in the tropical Indo-Pacific Ocean
CLOSEOmbai The Ombai Strait was artificially closed to block the propagation of the
Kelvin waves through the Ombai Strait
REMOVEIslands The Sumba and Savu/Dao Islands were artificially removed to reduce the loss
of Kelvin waves energy along the southern boundary of Sumba Island
WANG ET AL. 4 of 17
PACRUN) can be obtained by calculating the difference between IORUN1 and INDOPACRUN. IORUN2 and
PACRUN represent the forcing of the Indian Ocean and the Pacific Ocean, respectively. The local forcing on the
Timor currents can be obtained by calculating the difference between IORUN1 and IORUN2. The mean density
profile derived from the World Atlas 2013 version 2 (WOA13v2) was averaged between 10°S and 10°N in both
the Indian Ocean and Pacific Ocean, respectively, and was used for baroclinic mode decomposition. The model
has a realistic coastline geometry that is determined by the 200-m depth contour of the 1-min gridded global relief
data (ETOPO1) resolving the major exit straits of the ITF, including the Lombok Strait, the Ombai Strait, the
Timor Passage, and the Savu and Dao Straits. The LCSM was first forced in the tropical Indo-Pacific Ocean for
20 years using monthly climatological ERA5 wind stress data from 1979 to 2020. After a 20-year spin-up period,
all the runs were forced by the mean wind stress from the monthly ERA5 data from 2000 to 2020. The dynamics
of the LCSM result from the westward and downward propagation of baroclinic Rossby waves from the central
and eastern Pacific Ocean (Yang et al., 2020) and from the eastward and downward propagation of baroclinic
Kelvin waves along the equatorial and coastal Indian Ocean (Wang et al., 2020). The LCSM can effectively
reproduce the mean surface circulation of the tropical Indian Ocean and Pacific Ocean, including the Indonesian
seas using the first three baroclinic modes (Li et al., 2020; Wang et al., 2020). In this paper, we present only the
simulations from the first two baroclinic modes since the Timor transport is mainly controlled by the first two
vertical modes in the upper 400 m, as suggested by Potemra et al. (2003).
To investigate the specific propagation pathways of Kelvin waves in the Indian Ocean, additional two sensitivity
runs using the LCSM are performed (see Table 1): (a) closing the Ombai Strait artificially to block the propaga-
tion of the Kelvin wave along the Nusa Tenggara Island chain through the Ombai Strait (named CLOSEOmbai
run); and (b) removing the Sumba and Savu/Dao Islands to avoid partial energy propagation along the southern
boundary of the Sumba and Savu/Dao Islands to the Roti Island (named REMOVEIslands run). Both runs are
forced by the Indian Ocean winds in the west of 110°E to test the propagation of the equatorial and coastally
trapped Kelvin wave in the Indian Ocean.
3. Results
The mean structure in the vertical direction and the variability in the Timor current were analyzed using subsur-
face mooring data and historical measurements. And the transport through the Timor Passage was also estimated.
The seasonal and interannual variations in transport through the Timor Passage were further investigated.
3.1. Timor Current
The mooring-measured velocities were rotated to the along-strait direction based on the orientation of the pres-
sure gauges (Timor 66°T). Here, we defined positive values as indicating flows toward the Pacific Ocean side,
and negative values as indicating flows toward the Indian Ocean side. The time-depth plot of the along-strait
velocity (ASV) derived from the ADCP measurement shows unidirectional currents entering the Indian Ocean
through the Timor Passage (Figure 3b). The currents at 750, 1,000, and 1,400 m were also generally measured
as southwestward flows (Figure 3c). The currents below the thermocline have good correlations above the 90%
significance level, suggesting potential baroclinic processes. Strong seasonality in the upper layer can be clearly
identified. A stronger southwestward ASV is seen during the southeast monsoon (SEM, June-July-August), and
a weaker southwestward ASV is seen during the northwest monsoon (NWM, December-January-February;
Figure 3b). Semiannual variability below 200 m was also observed in 2018.
A linear interpolation method was adopted to obtain the currents among the ADCP and different current meters
at different depths since weak vertical velocity shear occurred below the thermocline. A unidirectional current
entering the Indian Ocean was found in the Timor Passage, with more intense along-passage flow near the surface
and less intense flow in lower layers. The maximum amplitude of the mean ASV was −42.8 cm s −1 at ∼60 m,
with a weakened flow in the middle layer and a slightly intensified flow near 1,400 m, which is consistent with
the study of Sprintall et al. (2009; Figure 4a). Based on the INSTANT mooring measurements, the currents in
the Timor Passage are mainly confined to the northern part of the passage (Sprintall et al., 2009; Figure 3c).
Therefore, a no-slip solid wall condition was used to estimate the velocity in the cross-strait section of the Timor
Passage (Figure 4b). The corresponding standard deviation of ASV was also estimated (Figure 4c).
WANG ET AL. 5 of 17
Figure 3. Time series of averaged zonal wind stress anomalies along the equator in the Indian (solid) and Pacific Oceans
(dotted). The zonal wind stress anomalies (dashed) and sea level anomalies (blue) in the Timor Passage are also shown (a),
along with the daily along-strait velocity (ASV) from the acoustic Doppler current profiler (ADCP) data (b), and the daily
ASV measured using the Aanderaa current meter at 750, 1,000, and 1,400 m (c). A 120-day low-pass filter was applied to the
data. The correlation coefficients of the currents at different depths are also labeled.
Stronger currents can be found in the upper 150 m, especially in April-July, over the 2-year measurement period
(Figure 3a), suggesting the strong seasonality in the Timor Passage currents in the passage. The ADCP meas-
urements have the best coverage in different seasons and, in combination with historical observations, facili-
tate an investigation of the seasonal variability in the Timor current. An empirical orthogonal function (EOF)
analysis also shows a two-layer structure for the first mode and second mode (Figure 5a). The first EOF mode
crosses zero at ∼200 m; this result is consistent with the zero crossings of the second baroclinic mode in tropical
oceans, further suggesting the involvement of baroclinic processes. Notably, these results are deeper than the
zero-crossing of 150 m simulated by Potemra et al. (2003). The first EOF mode dominates 49.0% of the variance,
and the second mode dominates 22.0% of the variance. These results suggest that the first EOF dominates the
variance in the Timor currents.
WANG ET AL. 6 of 17
Figure 4. Mean profile of the current (a) along-strait velocity (ASV) (solid curve) and its standard deviation (dotted curve), (b) the interpolated mean structure of the
ASV and its standard deviation (c) in the Timor Passage, using nonslip conditions at the coast and shelf. (d) ASV (solid curve) and its standard deviation (dotted curve),
(e) the interpolated mean structure of the ASV and its standard deviation (f) of the Timor Sill mooring.
WANG ET AL. 7 of 17
Figure 5. (a) The first (blue curve) and second (red curve) empirical orthogonal function (EOF) and (b) power spectral
density (PSD, cm 2 s −1) of the Timor along-strait velocity (ASV) at the mooring site at different depths.
The power spectrum of the ASV shows a stronger annual period in the upper 150 m and a clear semiannual period
below 200 m at the mooring site (Figure 5b). The intraseasonal variability amplitude is much weaker than the
seasonal variability in the upper 480 m. The harmonical analyses further confirmed stronger annual harmonics
of the ASV above 150 m with a maximum amplitude near the surface (Figure 6a). The strong westward currents
in June-September and weak currents in other seasons are consistent with the Asian-Australian monsoon system
in the Timor Passage. Although the amplitudes of semiannual harmonics are overwhelmed by the annual cycle
above 150 m, stronger semiannual harmonics are obvious near 50 m (with a maximum of 20 cm s −1 and below
200 m, with a maximum of 15 cm s −1 between 250 and 350 m). Opposite annual and semiannual harmonics are
found in the upper and lower 150 m and semiannual harmonics below 200 m are consistent with the semian-
nual cycles of the Wyrtki Jets in the equatorial Indian Ocean. Noticeably, the
maximum near 350 m first appears in October 2017 then propagates upward
close to 250 m in December 2017. Upward phase propagation is indicative of
downward energy propagation into the ocean interior, which is in agreement
with the structure of a Kelvin wave forced at the surface, as suggested by
linear wave theory (McCreary, 1984). In other words, upward phase propa-
gation signatured the downward penetration of Kelvin waves energy, which
is confirmed in the study of Drushka et al. (2010). These suggest a robust
pattern of downward energy propagation, associated with the baroclinic wave
propagation of the semiannual period in the Timor Passage. Correspond-
ing amplitude and phase of annual and semiannual harmonic analysis also
confirmed the consistent upward phase shift and downward energy propa-
gation (Figure S1 in Supporting Information S1). Waves propagate toward
the small phase (Wakata, 2001). Downward propagation of energy in the
annual harmonics only occurs in the upper 100 m (Figure S1a in Supporting
Information S1), suggesting that the annual cycle may be dominated by local
processes. Downward energy propagation for semiannual cycle can be seen
in the upper 480 m with stronger amplitudes below 200 m (Figure S1b in
Supporting Information S1).
Figure 6. (a) Annual-cycle and (b) semiannual-cycle harmonics of To validate the moored harmonics of the Timor currents, the harmonic anal-
along-strait velocity (ASV) at the mooring site in the Timor Passage in the
upper 500 m.
ysis of the Timor current and Ombai current from the INSTANT mooring
WANG ET AL. 8 of 17
Figure 7. (a) Annual-cycle and (b) semiannual-cycle harmonics of along-strait velocity (ASV) in the Timor Passage from the INSTANT measurement above 1,890 m
from January 2004 to December 2006. The annual-cycle (c) and semiannual-cycle (d) harmonics of ASV in the 500 m are magnified.
measurement are repeated for comparison (Figures 7 and 8). A consistent annual cycle is found in the upper
150 m, but opposite semiannual signals are found in the upper 500 m in both straits. Strong semiannual signals
in the depth ranges of 700–1,890 and 500–1,250 m in the Timor Passage can be seen clearly, which have been
explained by the deep-reaching Kelvin waves from the Ombai Strait (Molcard et al., 1996; Sprintall et al., 2009).
Semiannual signals also exist in the upper 500 m with an amplitude maximum at ∼200 m, which is much weaker
than the deeper current below 1,000 m in the Timor Passage (Figure 5d). The semiannual signals below 200 m
were not addressed thoroughly in the previous study. Opposite semiannual signals can be found above and below
500 m in the Timor Passage (Figures 7b and 7d). The phase lag between the INSTANT measurement and moor-
ing data may be due to the coarse resolution of the current meters at the Timor Sill. Coincidently, consistent
stronger semiannual signals in the Ombai Strait are found, with maxima in April-May and October-November
near 350 m and minima in January and July, resembling a semiannual signal (Figure 8). These findings are also
consistent with the newest mooring measurements. The possible relations are explained later in this paper using
a simple LCSM.
Figure 8. (a) Annual-cycle and (b) semiannual-cycle harmonics of along-strait velocity (ASV) in the Ombai Strait from the INSTANT measurement above 750 m from
August 2003 to December 2006.
WANG ET AL. 9 of 17
Figure 9. (a) Volume transport (Sv) of the Timor Passage using nonslip wall conditions in the different depth ranges: (black) 40–480 m, (red) 40–150 m, and (green)
150–480 m; (b) annual and semiannual harmonics of the volume transport. (c) The annual and semiannual cycles of sea level anomalies and alongshore wind stress
in the Timor Passage. (d) Ekman transport integrated from 12.2°S to 11°S along the longitude of the mooring longitude. A 120-day low-pass filter was applied to the
Ekman transport (smooth curve). Positive values indicate eastward Ekman transport associated with southwestward winds, which offsets the Indonesian Throughflow
(ITF) and negative values indicate westward Ekman transport associated with northwestward winds, which enhances the ITF.
3.2. Volume Transport Through the Timor Passage
The total transport was estimated by summing all the flows through the along-passage direction in the whole
water column. Here, we ignored the transport below 1,400 m considering the possible errors that arose due
to extrapolation since opposing currents were observed near the bottom in the calculations of the INSTANT
measurements (Sprintall et al., 2009). The gaps in the upper 40 m are set to a constant velocity equal to the
shallowest measured velocity. The mean transport through the Timor Passage in the upper 1,400 m is found to be
−9.9 ± 1.0 Sv using the no-slip boundary condition; this value is larger than the INSTANT results since reverse
flows were observed below 1,400 m (−7.5 Sv, Gordon et al., 2010; Sprintall et al., 2009). The mean ASV of
the Timor Sill mooring from the INSTANT measurement is observed to be directed toward the Pacific side below
1,400 m (Figure 4e), which reduces the westward transport to some extent. In addition, some errors inevitably
arise due to the use of single mooring data compared with the 3-year measurements derived from four moorings
deployed during the INSTANT program. The mean transport is −8.9 ± 3.3 Sv in the upper 480 m. The transport
values above and below 150 m were also calculated due to the different annual harmonics. The transport values
are 2.2 Sv between 40 and 150 m and 2.1 Sv between 150 and 480 m, suggesting that most of the ITF occurs in
the upper layer, especially above 150 m. For comparison, the total transport was also estimated using free-slip
boundary conditions. The corresponding transport through the Timor Passage is found to be 12.9 ± 5.0 Sv in the
upper 1,400 m, 10.8 ± 0.9 Sv in the upper 500 m, and 8.6 ± 2.7 Sv in the depth range of 150–500 m.
The time series of the transport through the Timor Passage in the upper 480 m was also estimated using the
no-slip boundary condition (Figure 9a). Seasonal variations are clearly seen in the upper 480 m (Figure 9b), with
stronger semiannual-cycle and weak annual-cycle harmonics of volume transport in the upper 480 m due to the
cancelation of the annual transport in the upper and lower layers. In the semiannual cycle, the maximum transport
occurs in March and September, and the minimum transport occurs in May and November, and this pattern is
related to the stronger semiannual cycles in the upper layer than the lower layer since the transport in the upper
150 m dominates the total transport in the upper 480 m. The annual cycle with maximum transport in boreal
summer-fall and minimum transport in boreal winter is also consistent with the seasonality of the ITF volume
transport proposed in earlier studies (Liu et al., 2015; Sprintall et al., 2009). The Ekman transport estimated
using the ASCAT daily wind stress data through the Timor Passage shows an annual cycle with consistency of
the mooring data (Figures 9b and 9d), suggesting that the Ekman transport dominates the annual cycle of Timor
transport. This finding also confirms that the annual cycle of Timor transport mainly occurs in the upper 150 m.
WANG ET AL. 10 of 17
Table 2 The annual cycles of sea level anomalies (SLA) and alongshore wind stress
Volume Transport (Standard Deviation) in 2018 and 2019 in Different are stronger than their semiannual cycles, suggesting that stronger annual
Depth Ranges cycles occur near the surface (Figure 9c). The semiannual harmonics of SLA
Year lead to alongshore wind stress for ∼1 month, further suggesting that the semi-
annual signals in the Timor Passage are not locally forced.
Depth (m) 2018 2019 Total
The strongest positive IOD event occurred in 2019, and the relationship
40–480 −4.40 (0.79) −4.45 (0.65) −4.36 (0.79)
between the volume transport through the Timor Passage and the remote
40–150 −2.18 (0.63) −2.33 (0.36) −2.22 (0.52)
wind forcings during the IOD event was also examined (Figure 3a). Pacific
150–480 −2.22 (0.69) −2.12 (0.68) −2.14 (0.68) westerly wind anomalies and Indian easterly wind anomalies were observed
during the positive IOD event in 2019; these anomalies forced upwelling
Kelvin waves in the Indian Ocean and upwelling Rossby waves in the Pacific
Ocean, and both depressed the dynamic height in the Timor region. The equatorial winds from the Indian Ocean
were much stronger than those in the Pacific Ocean, suggesting that the volume transport in the Timor Passage
was likely influenced by the Indian side. The volume transport in the Timor Passage was slightly larger in 2019
than in 2018 (Table 2 and Figure 9a). A clear decrease in local SLA in the Timor Passage is also consistent with
the zonal wind stress anomalies (Figure 3a). The correlation coefficients between the transport and wind stress
anomalies further suggest that the local wind stress dominated the volume transport through the Timor Passage
during the mooring measurement period (Table 3). Here, we show only the results of the mooring measurement
period in the upper 480 m. The longer time series of the total transport through the Timor Passage are also
discussed using reanalysis data and model data later in this paper.
4. Discussion
The seasonal variations in the Timor Passage are forced both locally by the monsoon (Meyers et al., 1995; Qu
et al., 1994) and remotely by signals that propagate from the equatorial Pacific and Indian Oceans (Clarke &
Liu, 1993; Meyers, 1996; Molcard et al., 1996; Potemra, 2001; Potemra et al., 2002; Wijffels & Meyers, 2004).
The westward propagation of Rossby waves from the Pacific Ocean and eastward propagation of Kelvin waves
from the Indian Ocean may play important roles, as was suggested by Wijffels et al. (2004).
The annual harmonics of the alongshore wind stress exhibited a significant correlation with the ASV in the upper
150 m above the 95% confidence level, suggesting that the local monsoonal winds dominate the annual cycle in
the Timor Passage. The semiannual harmonics of the ASV below 200 m exhibit consistent semiannual variability
in the Wyrtki Jets in the Indian Ocean with maxima in May and November during the monsoon transition time
(Nagura & McPhaden, 2010), suggesting a possible important influence from the equatorial Indian Ocean on the
Timor Passage. The westward propagation of the Rossby wave from the Pacific Ocean may also play an important
role, as suggested by Wijffels et al. (2004). The significant correlation between the ASV in the Timor Passage
and the equatorial wind stress in the Indian Ocean and Pacific Ocean suggests that the semiannual harmonics
of the ASV are possibly controlled by remote forcings, which is validated using the LCSM later in this section.
In this study, we use a linear model to investigate vertical propagation of remotely (the Indian Ocean and Pacific
Ocean) and locally forced baroclinic Kelvin waves and to further quantify the local and remote wind forcings on
the annual and semiannual harmonics of the current and transport through the Timor Passage. Three sensitivity
experiments were conducted using a simple LCSM. In this way, the Indian and Pacific wind-forced annual and
semiannual harmonics of the ASV in the Timor Passage can be obtained. The
difference between IORUN1 and IORUN2 suggests the influence of the local
Table 3 forcing on the vertical structure of the annual cycle in comparison with the
Correlation Coefficients Between the Wind and Transport in Different influence of the Indian Ocean and the Pacific Ocean (Figure 10). The results
Depth Ranges from the local forcing demonstrate that the consistent annual harmonics of
Forcing the ASV in the upper 150 m were reproduced reasonably well, with a maxi-
mum near the surface. The annual ASV harmonics derived from IORUN1
Depth (m) Local wind IO PO
and PACRUN are different from the mooring data, further suggesting the
40–480 −0.24 −0.21 −0.07 dominant role of the local monsoonal winds.
40–150 −0.69 −0.27 −0.16
The annual harmonics of the ASV can be obtained fairly well by the local
150–480 0.24 −0.04 0.04 wind forcing, suggesting that the annual cycle of the Timor current is mainly
driven by the local forcing (Figure 10a). The semiannual harmonics of the

ASV simulated by IORUN2 below 200 m can also be reproduced reasonably
well, although the amplitudes are smaller in comparison with the moored
currents (Figure 11b). An upward phase shift can be seen clearly suggesting
the downward propagation of the energy. However, a similar structure of the
semiannual signals cannot be simulated by PACRUN (not shown), suggest-
ing that the forcing is unlikely to originate from the Pacific side. Similarly,
the local wind forcing also cannot reproduce the semiannual signals of Timor
transport with weak semiannual signals (Figure 10b). Therefore, the semian-
nual harmonics of the ASV below 200 m are mainly controlled by the remote
Indian side (Figure 12a).
The transport harmonics through the Timor Passage were also esti-
mated from different forcings based on the above sensitivity experiments
(Figure 12). The total transport through the Timor Passage is not dominated
by the annual cycle due to the cancelation of the annual transport mutually
forced by the Indian Ocean, Pacific Ocean, and local wind forcing in the
Figure 10. Linear continuously stratified model (LCSM)-simulated upper 480 m (Figure 12). The annual cycle of transport is largely reproduced
annual-cycle harmonics (a) and semiannual-cycle harmonics (b) of the by the locally forced model simulation although there is a lead and lag time
along-strait velocity (ASV) at the mooring site in the upper 500 m of the between local-forced transport and mooring measurements due to the selec-
Timor Passage forced by local winds. tion of the local forcing region (Figure 12c). Consistencies among the local
forced model, Ekman transport, and mooring measurements further suggest
that the annual cycles of Timor transport are mainly dominated by Ekman
transport (Figures 9b, 9d, and Figure 12c). However, strong annual harmonics of the Timor transport simulated
by the Indian Ocean forcing and Pacific Ocean forcing alone are not consistent with the moored observations.
Therefore, the annual transport through the Timor Passage is mainly driven by local forcing. Consistent semian-
nual harmonics of the Timor transport were simulated in the upper 500 m, which are mainly forced by the Indian
side (Figures 12a and 9b).
The different time scales of the total transport in the Timor Passage are the result of remote forcing from both
the Pacific and Indian Ocean winds and local forcing generated within the regional Indonesian seas (Schiller
et al., 2010; Sprintall et al., 2009). To further investigate the forcing mechanism of the Timor Passage transport,
a long time series of interannual variability in transport is discussed using the BRAN2020 data and simple
LCSM. The interannual transport anomalies through the Timor Passage are estimated in different depth ranges
(Figure 13). The Timor transport in three different depth ranges increased
during the mooring measurement periods. The transport anomalies in 2019
are not obvious since more transport occurs through the Lombok and Ombai
Straits to the Indian Ocean due to the topography boundary and model reso-
lution (Schiller et al., 2010). The IOD and ENSO are important anomalous
climatic events in the Indian Ocean and Pacific Ocean, which can be defined
by the Niño3 index and Dipole Mode Index (DMI; Figure 13b). To study
the influence of the IOD and ENSO on the Timor transport, the correlation
coefficients between the Timor transport in different depth ranges and Niño3
and DMI are estimated. The correlation analysis suggests that the remote
forcing from the Pacific Ocean has more significant influences on the Timor
Passage, especially in the deep layer (the total transport and the transport in
the upper 500 m) above the 90% confidence level. The Indian Ocean also
plays a secondary role. In other words, the Pacific forcing dominates the
interannual variability especially when ENSO and IOD events cooccur.
To separate the role of the Indian Ocean and Pacific Ocean forcings on trans-
port through the Timor Passage, the interannual variability in Timor trans-
port forced by different regions is also calculated using the simple LCSM
Figure 11. Linear continuously stratified model (LCSM)-simulated
(Figure 13c). The correlation coefficients between the total transport and the
annual-cycle harmonics (a) and semiannual-cycle harmonics (b) of the
along-strait velocity (ASV) at the mooring site in the upper 500 m of the transport forced by different regions suggest that the Pacific forcing domi-
Timor Passage forced by the remote Indian winds. nates the interannual variability of the Timor transport with a correlation
D,QGLDQ)RUFLQJ+DUPRQLFV coefficient of 0.95, and the Indian Ocean forcing also greatly influences the

Timor transport largely with a correlation coefficient of 0.65 above the 90%

confidence level. Strong positive and negative IOD events occurred in 2019

6Y
$QQ
and 2016, respectively, based on the time series of the DMI (Figure 13b,

6HPL red). Simultaneously, no large-scale anomalous climate events occurred on

6 1 0 0 - 6 1 0 0 - 6 1 the Pacific Ocean side. Based on the LCSM simulations, the Indian Ocean
E3DFLILF)RUFLQJ+DUPRQLFV wind-forced transport dominated the interannual variability of the total trans-
port in 2016 and 2019 (red and black) and the Pacific winds forcing had little
effect on the Timor transport (blue and black). That is, IOD events domi-

6Y
nate the interannual variability of Timor transport when an ENSO event is

not occurring simultaneously in the Pacific Ocean. Therefore, the increased

6 1 0 0 - 6 1 0 0 - 6 1 transport through the Timor Passage observed in the mooring measurements
F/RFDO)RUFLQJ+DUPRQLFV in 2019 was mainly driven by the Indian Ocean forcing. The Pacific forcing
also plays a minor role, and local forcing also decreases the Timor Passage
transport to some extent. The correlation between local wind stress and

6Y
Timor transport in the upper 150 m from BRAN2020 data suggests that the
local wind stress mainly influences the transport in the upper 150 m with a

6 1 0 0 - 6 1 0 0 - 6 1 correlation coefficient of −0.42 above the 95% confidence level. The moor-
ing measurements confirmed the above result.
Figure 12. Annual (solid) and semiannual (dashed) volume transport
There are two possible pathways of Kelvin waves propagating from the
harmonics in the Timor Passage derived from the IORUN2 (a), PACRUN (b),
and local forcing (c) simulations. equatorial Indian Ocean to the Timor Passage: Pathway I: equatorial and
Figure 13. (a) Total transport (red) and transport in the upper 150 m (blue) and 500 m (green) through the Timor Passage of
the BRAN2020 from 1993 to 2021. The black box shows the transport during the mooring measurement period (September
2017 to December 2019). (b) The Niño 3 index (blue) and DMI from 1993 to 2021 (red); (c) Linear continuously stratified
model (LCSM)-simulated transport through the Timor Passage forced by the Indo-Pacific (black), Indian (red), Pacific (blue),
and local winds (green). Positive and negative values indicate decreases and increases in the transport through the Timor
Passage, respectively.
Figure 14. Hovmöller plots of seasonal sea level anomalies averaged within 1° along the Pathway I (a) and Pathway II (b). The abbreviation EB” indicates the eastern
boundary of the Indian Ocean. The letters “L,” “O,” and “T” indicate the location of the Lombok, Ombai Straits, and Timor Passage.
coastal-trapped Kelvin waves propagate along Nusa Tenggara Island to the Ombai Strait and then recirculate
along the northern boundary of the Timor Island, and bypass Roti Island to the mooring location; Pathway II:
equatorial and coastal-trapped Kelvin waves propagate along the southern boundary of the Sumba Island and then
pass the Savu/Dao Islands to the Timor Passage (Schiller et al., 2010; Sprintall et al., 2009). Seasonal variations
of SLA along the two pathways are calculated (Figure 14). The annual oscillation in the central Indian Ocean and
the semiannual oscillation in the eastern Indian Ocean of sea level can be easily found, which is consistent with
previous studies (Wang et al., 2017, 2018). The propagating annual signals can be seen from the east boundary of
the Indian Ocean to the Lombok, Ombai, and Timor along the Pathway I instead of Pathway II. Since semiannual
harmonics in the Timor Passage mainly occur below 200 m, it is hard to be recognized from the surface sea level
observations. In addition, it is difficult to distinguish the proportion of the two possible propagating pathways of
the Kelvin wave from the Nusa Tenggara Island to the Timor Island based on the sea level observations only due
to the errors of SLA near the coast. To further investigate the specific propagation pathways of the Kelvin waves,
two sensitivity runs using the LCSM are performed (a) closing the Ombai Strait artificially to block the propaga-
tion of the Kelvin wave through the Ombai Strait (named CLOSEOmbai run); and (b) removing the Sumba and
Savu/Dao Islands to avoid partial energy propagation along the southern boundary of the Sumba and Savu/Dao
Islands to the Roti Islands (named REMOVEIslands run). In this way, the propagating pathways of Kelvin waves
and associated energy distribution can be estimated. The numerical results show that the semiannual harmonics
of the Timor current increased slightly when the Ombai Strait was artificially closed (Figure 15, left panel).
That is, most of the energy propagates along the Pathway I, and little energy dissipates through the Ombai Strait
(Figure 1, lower panel dashed). The semiannual harmonics of the Timor current are almost unchanged when the
Sumba and Savu/Dao Islands are removed artificially (Figure 15, right panel), suggesting that the energy mainly
Figure 15. Left panel: annual (a) and semiannual (b) harmonics of the Timor current forced by the Indian winds when the Ombai Strait was artificially closed; right
panel: annual (c) and semiannual (d) harmonics of the Timor current forced by the Indian winds when Sumba and Savu/Dao Islands were artificially removed. The
land-sea distributions are also overlaid in boxes (a, c).
propagates along Pathway I instead of Pathway II. The propagating pathways of the energy are shown in Figure 1.
Most of the semiannual-signal energy propagates along the solid curve and less energy propagates along the Nusa
Tenggara Island into the Ombai Strait and along the southern boundary of the Sumba Island.
5. Conclusions
The transport and variability of the Timor Passage currents are investigated in this study using data collected
over 2 years by a subsurface mooring deployed at the center of the passage (122.88°E, 11.26°S) from September
2017 to December 2019. The mean currents are found to flow westward into the Indian Ocean at all depths, with
a velocity maximum core near 50 m. The EOF analysis shows the two-layer structure associated with the baro-
clinic processes in the Timor Passage. The ASV features strong annual harmonics in the upper 150 m and weak
annual harmonics below 200 m largely associated with local monsoonal forcing in the Timor Passage instead of
the Indian or Pacific forcing. Stronger semiannual harmonics rather than annual cycles below 200 m are revealed
with upward phase shifts associated with remote Kelvin waves from the Indian Ocean. Forcing from the Indian
Ocean side contributes more to the vertical structure of the semiannual cycle through the Ombai Strait into the
Timor Passage.
The mean volume transport through the Timor Passage in the upper 1,400 m is estimated to be −9.9 ± 1.0 Sv
(1 Sv = 10 6 m 3 s −1) into the Indian Ocean, assuming the no-slip sidewall condition; this value can be compared
with the value of −7.5 Sv derived from INSTANT measurements from the surface to the bottom of the Timor
Passage. The transport in the upper 480 m is −4.36 Sv, suggesting that a large portion of the ITF transport through
the Timor Passage occurs in the upper layer. The transport variability is also found to be dominated by the semi-
annual cycle rather than the annual cycle in the upper 480 m, which is likely associated with the semiannual
cycles in the Indian Ocean. This finding is also consistent with the ASV below 200 m. The transport through
the Timor Passage during a positive IOD event is slightly larger than that during a normal year, especially in
the upper 150 m, further suggesting that the influence of the IOD on the interannual variability in the transport
may mainly occur in the upper layer based on the mooring measurements. A longer time series of transport from
reanalysis and linear model data shows that the Pacific forcing associated with the ENSO events dominates the
interannual variability in Timor transport when the ENSO and IOD events cooccur, and the Indian Ocean forcing
associated with the IOD events dominates the interannual variability in Timor transport when no ENSO events
occur in the Pacific Ocean.
Data Availability Statement

The moored ADCP data collected in the Timor Passage will be free to download at the following website: http://
dx.doi.org/10.12157/IOCAS.20220329.001. The subsurface mooring data derived from the INSTANT program
can be downloaded at the following site: http://www.marine.csiro.au/∼cow074/instantdata.htm. The topography
data can be downloaded at the following site: https://www.ngdc.noaa.gov/mgg/global. The daily scatterometer
wind stress data from the advanced scatterometer (ASCAT) can be downloaded from the Asia-Pacific Data
Research Center (APRC) project (http://apdrc.soest.hawaii.edu/data/data.php). The ERA5 wind stress data can
be downloaded at http://apdrc.soest.hawaii.edu/erddap/griddap/hawaii_soest_d124_2bb9_c935.html. The Blue-
link Reanalysis (BRAN2020) data can be accessed at https://research.csiro.au/bluelink/outputs/data-access/. The
interactive data language (IDL) and MATLAB were used to generate all figures.
Acknowledgments References
This work was financially supported
by the Natural Science Foundation Bentamy, A., & Fillon, D. C. (2012). Gridded surface wind fields from Metop/ASCAT measurements. International Journal of Remote Sensing,
of China (41720104008), the MNR 33(6), 1729–1754. https://doi.org/10.1080/01431161.2011.600348
Program on Global Change and Air-Sea Clarke, A. J., & Liu, X. (1993). Observations and dynamics of semiannual and annual sea levels near the Eastern Equatorial Indian-Ocean Bound-
interactions (GASI-04-WLHY-03), the ary. Journal of Physical Oceanography, 23(2), 386–399. https://doi.org/10.1175/1520-0485(1993)023<0386:oadosa>2.0.co;2
Strategic Priority Research Program of Cowley, R., Heaney, B., Wijffels, S., Pender, L., Sprintall, J., Kawamoto, S., & Molcard, R. (2008). INSTANT Sunda data report description and
Chinese Academy of Sciences (Grant quality control. Retrieved from http://www.marine.csiro.au/_cow074/INSTANTdataQC_v4.pdf
XDB42000000), the Laoshan Laboratory Doi, T., Behera, S. K., & Yamagata, T. (2020). Predictability of the super IOD event in 2019 and its link with El Niño Modoki. Geophysical
(2022QLSL010304-4), the Natural Research Letters, 47, e2019GL086713. https://doi.org/10.1029/2019GL086713
Science Foundation of China (NSFC) Drushka, K., Sprintall, J., Gille, S. T., & Brodjonegoro, I. (2010). Vertical structure of Kelvin waves in the Indonesian Throughflow exit passages.
(41776011), and the National Key R&D Journal of Physical Oceanography, 40(9), 1965–1987. https://doi.org/10.1175/2010jpo4380.1
Program of China (2019YFA0606702). Ducet, N., Le Traon, P. Y., & Reverdin, G. (2000). Global high-resolution mapping of ocean circulation from TOPEX/Poseidon and ERS-1 and
D. Yuan is supported by the “Taishan -2. Journal of Geophysical Research, 105(C8), 19477–19498. https://doi.org/10.1029/2000JC900063
Scholar” Program and the “Kunpeng Gordon, A. L., Sprintall, J., Van Aken, H. M., Susanto, D., Wijffels, S., Molcard, R., et al. (2010). The Indonesian Throughflow during 2004–2006 as
Outstanding Scholar” Program of the observed by the INSTANT program. Dynamics of Atmospheres and Oceans, 50(2), 115–128. https://doi.org/10.1016/j.dynatmoce.2009.12.002
FIO/MNR of China. Hautala, S. L., Sprintall, J., Potemra, J. T., Chong, J. C., Pandoe, W., Bray, N., & Ilahude, A. G. (2001). Velocity structure and transport of the
Indonesian Throughflow in the major straits restricting flow into the Indian Ocean. Journal of Geophysical Research, 106(C9), 19527–19546.
https://doi.org/10.1029/2000JC000577
Li, X., Yang, Y., Li, R., Zhang, L. L., & Yuan, D. L. (2020). Structure and dynamics of the Pacific North Equatorial Subsurface Current. Scientific
Report, 10(1), 11758. https://doi.org/10.1038/s41598-020-68605-y
Liu, Q.-Y., Feng, M., Wang, D., & Wijffels, S. (2015). Interannual variability of the Indonesian Throughflow transport: A revisit based on 30
year expendable bathythermograph data. Journal of Geophysical Research: Oceans, 120, 8270–8282. https://doi.org/10.1002/2015JC011351
Masumoto, Y., & Yamagata, T. (1996). Seasonal variations of the Indonesian Throughflow in a general ocean circulation model. Journal of
Geophysical Research, 101(C5), 12287–12293. https://doi.org/10.1029/95JC03870
McCreary, J. P. (1984). Equatorial beams. Journal of Marine Research, 42(2), 395–430.
Meyers, G. (1996). Variation of Indonesian Throughflow and the El Niño-Southern Oscillation. Journal of Geophysical Research, 101(C5),
12255–12263. https://doi.org/10.1029/95JC03729
Meyers, G., Bailey, R. J., & Worby, A. P. (1995). Geostrophic transport of Indonesian Throughflow. Deep Sea Research Part I: Oceanographic
Research Papers, 42(7), 1163–1174. https://doi.org/10.1016/0967-0637(95)00037-7
Molcard, R., Fieux, M., & Ilahude, A. G. (1996). The Indo-Pacific throughflow in the Timor Passage. Journal of Geophysical Research, 101(C5),
12411–12420. https://doi.org/10.1029/95JC03565
Molcard, R., Fieux, M., Swallow, J. C., Ilahude, A. G., & Banjarnahor, J. (1994). Low-frequency variability of the currents in Indonesian Chan-
nels (Savu-Roti and Roti-Ashmore Reef). Deep Sea Research Part I: Oceanographic Research Papers, 41(11–12), 1643–1661. https://doi.
org/10.1016/0967-0637(94)90066-3
Molcard, R., Fieux, M., & Syamsudin, F. (2001). The throughflow within Ombai Strait. Deep Sea Research Part I: Oceanographic Research
Papers, 48(5), 1237–1253. https://doi.org/10.1016/s0967-0637(00)00084-4
Nagura, M., & McPhaden, M. J. (2010). Wyrtki Jet dynamics: Seasonal variability. Journal of Geophysical Research: Oceans, 115(C7). https://
doi.org/10.1029/2009JC005922
Peña-Molino, B., Sloyan, B. M., Nikurashin, M., Richet, O., & Wijffels, S. E. (2022). Revisiting the seasonal cycle of the Timor Throughflow:
Impacts of winds, waves and eddies. Journal of Geophysical Research: Oceans, 127, e2021JC018133. https://doi.org/10.1029/2021JC018133
Potemra, J. T. (2001). Contribution of equatorial Pacific winds to southern tropical Indian Ocean Rossby waves. Journal of Physical Oceanogra-
phy, 106(C2), 2407–2422. https://doi.org/10.1029/1999JC000031
Potemra, J. T., Hautala, S. L., & Sprintall, J. (2003). Vertical structure of Indonesian Throughflow in a large-scale model. Deep Sea Research Part
II: Topical Studies in Oceanography, 50(12–13), 2143–2161. https://doi.org/10.1016/s0967-0645(03)00050-x
Potemra, J. T., Hautala, S. L., Sprintall, J., & Pandoe, W. (2002). Interaction between the Indonesian seas and the Indian Ocean in observations and
numerical models. Journal of Physical Oceanography, 32(6), 1838–1854. https://doi.org/10.1175/1520-0485(2002)032<1838:ibtisa>2.0.co;2
Qu, T., Meyers, G., Godfrey, J. S., & Hu, D. (1994). Ocean dynamics in the region between Australia and Indonesia and its influence on the
variation of sea surface temperature in a global general circulation model. Journal of Physical Oceanography, 99(C9), 18433–18445. https://
doi.org/10.1029/94JC00858
Schiller, A., Wijffels, S. E., & Sprintall, J. (2007). Chapter 8 variability of the Indonesian Throughflow: A review and model-to-data comparison.
In H. Kawahata, & Y. Awaya (Eds.), Elsevier oceanography series (pp. 175–494). Elsevier. https://doi.org/10.1016/S0422-9894(06)73008-2
Schiller, A., Wijffels, S. E., Sprintall, J., Molcard, R., & Oke, P. R. (2010). Pathways of intraseasonal variability in the Indonesian Throughflow
region. Dynamics of Atmospheres and Oceans, 50(2), 174–200. https://doi.org/10.1016/j.dynatmoce.2010.02.003
Sprintall, J., & Révelard, A. (2014). The Indonesian Throughflow response to Indo-Pacific climate variability. Journal of Geophysical Research:
Oceans, 119, 1161–1175. https://doi.org/10.1002/2013JC009533
Sprintall, J., Wijffels, S. E., Molcard, R., & Jaya, I. (2009). Direct estimates of the Indonesian Throughflow entering the Indian Ocean: 2004–2006.
Journal of Geophysical Research, 114, C07001. https://doi.org/10.1029/2008JC005257
Thompson, R. O. R. Y. (1983). Low-pass filters to suppress inertial and tidal frequencies. Journal of Physical Oceanography, 13(6), 1077–1083.
https://doi.org/10.1175/1520-0485(1983)013<1077:lpftsi>2.0.co;2;2
Van Riehl, P. M. (1943). Introductory remarks and oxygen content. In The Snellius expedition, oceanographic results, Part V: The bottom water
(Vol. II, p. 77). E. J. Brill.
Wakata, Y. (2001). Kelvin waves excited by intraseasonal wind forcing. Journal of Oceanography, 57, 693–707. https://doi.
org/10.1023/a:1021232408589
Wang, J., Yuan, D., Li, X., Li, Y., Wang, Z., Hu, X., et al. (2020). Moored observations of the Savu Strait Currents in the Indonesian seas. Journal
of Geophysical Research: Oceans, 125, e2020JC016082. https://doi.org/10.1029/2020JC016082
Wang, J., Yuan, D., & Zhao, X. (2017). Impacts of Indonesian Throughflow on seasonal circulation in the equatorial Indian Ocean. Chinese
Journal of Oceanology and Limnology, 35(6), 1261–1274. https://doi.org/10.1007/s00343-017-6196-0
Wang, J., Yuan, D., & Zhao, X. (2018). Influence of western boundary reflection on seasonal circulation in the equatorial Indian Ocean. Journal
of Oceanography, 74(3), 327–338. https://doi.org/10.1007/s10872-018-0463-3
Wang, J., Zhang, S., Jiang, H., & Yuan, D. (2022). Effects of 2019 subsurface Indian Ocean initialization on the forecast of the 2020/2021 La Niña
event. Climate Dynamics. https://doi.org/10.1007/s00382-022-06442-7
Wijffels, S., & Meyers, G. (2004). An intersection of oceanic waveguides: Variability in the Indonesian Throughflow region. Journal of Physical
Oceanography, 34(5), 1232–1253. https://doi.org/10.1175/1520-0485(2004)034<1232:aioowv>2.0.co;2;2
Yang, Y., Li, X., Wang, J., & Yuan, D. (2020). Seasonal variability and dynamics of the Pacific North Equatorial subsurface current. Journal of
Physical Oceanography, 50(9), 2457–2474. https://doi.org/10.1175/jpo-d-19-0261.1

Wang Etal JGR2022

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Wang Etal JGR2022

Uploaded by

Copyright:

Available Formats

21699291, 2022, 11, Downloaded from https://agupubs.onlinelibrary.wiley.com/doi/10.

2. Data and Methods

2.1.2. Other Data Sets

2.2. Ocean Models

3.1. Timor Current

3.2. Volume Transport Through the Timor Passage

driven by the local forcing (Figure 10a). The semiannual harmonics of the

nate the interannual variability of Timor transport when an ENSO event is

Data Availability Statement

You might also like