Pathways of intraseasonal Kelvin waves in the Indonesian Throughflow regions derived from satellite altimeter observation

XU Teng-Fei， WEI Ze-Xun， CAO Guo-Jiaoand LI Shu-Jiang

aFirst Institute of Oceanography， State Oceanic Administration， Qingdao， China；bLaboratory for Regional Oceanography and Numerical Modeling， Qingdao National Laboratory for Marine Science and Technology， Qingdao， China；cKey Laboratory of Marine Science and Numerical Modeling， State Oceanic Administration， Qingdao， China

Pathways of intraseasonal Kelvin waves in the Indonesian Throughflow regions derived from satellite altimeter observation

XU Teng-Feia，b，c， WEI Ze-Xuna，b，c， CAO Guo-Jiaoa，b，cand LI Shu-Jianga，b，c

aFirst Institute of Oceanography， State Oceanic Administration， Qingdao， China；bLaboratory for Regional Oceanography and Numerical Modeling， Qingdao National Laboratory for Marine Science and Technology， Qingdao， China；cKey Laboratory of Marine Science and Numerical Modeling， State Oceanic Administration， Qingdao， China

The gridded sea level anomaly （SLA） data-set provided by AVISO is used to track the propagation of intraseasonal Kelvin waves in the Indonesian Throughfow （ITF） region. The large root mean square of intraseasonal SLA along the Sumatra and Java coast is closely related to the propagation of intraseasonal Kelvin waves that derive from the equatorial Indian Ocean. These Kelvin waves are further found to propagate following diferent pathways at the Lombok Strait. Pathway A propagates eastward throughout the Sumba Strait and Savu Sea to reach the Ombai Strait. Pathway B penetrates into Lombok and propagates northward to reach the Makassar Strait. Pathway C propagates southeastward along the southwest coast of the Sumba Island. The equatorial Kelvin waves take around 15 days to travel from the equatorial Indian Ocean to Lombok Strait， and around 5 days to penetrate into the Makassar and Ombai straits. The Kelvin wave-induced SLA persists in the ITF region for an additional 5 days and then diminishes subsequently. The phase speeds of these intraseasonal Kelvin waves along Pathways A， B， and C are 1.91-2.86， 1.69， and 1.96 m s-1，respectively-in agreement with the frst two baroclinic modes of Kelvin waves.

ARTICLE HISTORY

Revised 9 April 2016

Accepted 27 April 2016

Indonesian Throughfow；Kelvin waves； intraseasonal variability； pathway；waveguide

利用卫星高度计资料，研究了季节内Kelvin波在印度尼西亚贯穿流海域的传播。起源于赤道印度洋的Kelvin波抵达印度洋东边界后，以沿岸Kelvin波的形式沿苏门答腊-爪哇岛西南沿岸向东传播，15天后抵达龙目海峡，并分为三支：一支向北传播， 5天后抵达望加锡海峡，相速度约为1.69 m s-1；一支向东传播，5天后抵达翁拜海峡，相速度约为1.91-2.86 m s-1；一支沿松巴岛西南沿岸传播，相速度约为1.96 m s-1。季节内Kelvin波在该海域的传播速度与第一和第二斜压模Kelvin波一致。

1. Introduction

The Indonesian Throughfow （ITF）， which fows from the equatorial Pacifc into the Indian Ocean through the Indonesian seas， provides the only oceanic channel for inter-ocean exchange in the tropical oceans. The importance of the ITF lies in the fact that it is the key component of the so-called ‘ocean conveyor belt’， which plays an important role in the heat and salt balance of the global ocean （Gordon 1986； Broeker 1991）. In addition， the ITF also serves as a unique oceanic signal channel， through which tropical Indian Ocean anomalies can propagate to the equatorial Pacifc Ocean to infuence ENSO （Yuan et al. 2011； Yuan， Zhou， and Zhao 2013）. Previous studies have suggested signifcant intraseasonal oceanic variability （ISV） in the main infow and outfow straits and passages of the ITF， i.e. the Lombok Strait， Ombai Strait， Timor Passage， and Makassar Strait， based on in situ observations（Arief and Murray 1996； Ffeld and Gordon 1996； Molcard，Fieux， and Ilahude 1996； Chong et al. 2000； Wijfels and Meyers 2004） and satellite data （Syamsudin， Kaneko， and Haidvogel 2004； Iskandar et al. 2005）. These ISVs can be traced back to the intraseasonal equatorial Kelvin waves forced by the MJO over the equatorial central Indian Ocean（Qiu， Mao， and Kashino 1999； Schiller et al. 2010）.

Beneft from the International Nusantara Stratifcation and Transport 2004-2006 program （Sprintall et al. 2004），the ISV of the ITF in the Indonesian seas has been further revealed （Sprintall et al. 2009； Gordon et al. 2010）. Pujiana et al. （2009） suggested that the dominant variability of 45-90 days in the Makassar Strait thermocline is due to the combination of intraseasonal waves propagating from the Lombok Strait and Sulawesi Sea， rather than exertedby the local atmospheric ISV forcing. The ISV in Makassar is attributed to the second baroclinic mode of Kelvin waves propagating from the Lombok Strait along the 100-m isobaths， which is estimated to reduce the southward transport through the Makassar Strait by up to 2 Sv （Pujiana，Gordon， and Sprintall 2013）. The along-channel fows in Lombok and Ombai have also been found to be subject to ISV， which explains the fow reversal during the monsoon transition period （Sprintall et al. 2009； Gordon et al. 2010）. The ISVs in the outfow straits have been identifed to be associated with the frst and second baroclinic Kelvin waves that derive from the equatorial Indian Ocean， which propagate along the equator and the southwest coast of the Lesser Sunda Islands （Schiller et al. 2010； Iskandar et al. 2014）.

As supplements to in situ observations， satellite altimeter products provide high-resolution sea level anomaly （SLA） data， not only in the straits but over the entire ITF region， making it possible to track the propagation of intraseasonal Kelvin waves in the ITF region. Drushka et al. （2010） and Pujiana， Gordon， and Sprintall （2013） used weekly altimeter SLA data as evidence of the propagation of Kelvin waves. However， the pathways of intraseasonal Kelvin waves that derive from the equatorial Indian Ocean in the ITF region-in particular after being trapped by the eastern boundary-have not been examined using satellite altimeter observations. In this study， SLA data obtained from AVISO are used to describe the diferent pathways of the intraseasonal Kelvin waves in the ITF region.

2. Data and methodology

This study employs the gridded SLA products （Version 5.0）produced by Segment Sol multi-missions dALTimetrie，d’orbitographie et de localisation précise/Data Unifcation and Altimeter Combination System （SSALTO/DUACS） and distributed by AVISO， with support from CNES （http:// www.aviso.altimetry.fr/duacs/）. The data-set is daily， with a resolution of 0.25° × 0.25°， available from October 1992 to the present day. The sea surface winds from ERA-Interim，which covers the period 1979 to the present day， with a horizontal resolution of 0.75° × 0.75° and time interval of 6 h （Dee et al. 2011）， are also used to show the generation and propagation of wind-forced Kelvin waves.

The SLA and sea surface wind are fltered by a 20-90-day band-pass flter， which are defned as intraseasonal anomalies in the following text. Positive intraseasonal events are defned as the peak of positive intraseasonal SLA greater than 1 standard deviation. The composite analysis is achieved by averaging each positive event in a time window of ±35 days. The lag correlations are calculated based on the band-passed anomalies， with the Student’s t-test used to represent the level of signifcance.

3. Results

The root mean square （RMS） of the intraseasonally bandpassed SLA in the ITF region shows strong variation along the Sumatra and Java coast， with the largest RMS beyond 5 cm （Figure 1）. These ISVs are attributable to the propagation of intraseasonal Kelvin waves along the coast that derive from the equatorial Indian Ocean. Therefore， we defne some key areas and straits or passages to study the ISVs and their propagation in the ITF region. It should be noted that there is also a strong RMS in the southeastern Indian Ocean between 10 and 12°S， which is related to mesoscale eddies induced by the baroclinic instability of the South Equatorial Current （Feng and Wijfels 2002； Yang et al. 2015）. The dynamics of these ISVs is beyond the scope of this paper because it is independent of the propagation of Kelvin waves.

Figure 1.RMS of intraseasonal SLAs （20-90-day band-pass fltered） in the ITF region.

Figure 2.Composite phase of ISV in diferent straits and passages of the ITF.

Figure 2 shows the composite phase of positive intraseasonal events in each key area shown in Figure 1 using 20-90-day band-passed SLAs. The averaged period of ISVs along west coast of the Sumatra and in all straits isaround 25-30 days. The amplitude at the west coast of the Sumatra is 4 cm. The amplitudes in other areas varies from 5.5 cm south of Sunda Strait to 1.5 cm at Timor Passage， suggesting that the Kelvin waves have decayed during propagation， which may be attributable to the fact that the equatorial Kelvin waves are partly penetrating into each strait subsequently， and partly propagating along the southwest coast of the Lesser Sunda Islands during the propagation.

Figure 3 shows snapshots of the intraseasonal SLA and sea surface wind anomalies for the composite phase shown in Figure 2. Since the Kelvin waves bifurcate at Lombok Strait，we therefore defne the ISVs of SLA at Lombok as the reference to show the propagation of intraseasonal Kelvin waves from the equatorial Indian Ocean to the Indonesian seas. The results using the ISVs in other straits as the reference show similar patterns and are thus omitted here. Day（0）indicates the peak of the intraseasonal SLA at Lombok Strait. On day（-20）， before the Lombok peak， there are westerlies over the equatorial Indian Ocean that drive downwelling Kelvin waves to induce positive SLAs， which propagate eastward to reach western Sumatra on day（-15） （Figure 3（a）and （b））. Between day（-20） and day（+15）， there are easterlies and negative SLAs in the Indonesian seas and south of Java. The equatorial Kelvin waves are trapped by the eastern boundary and then propagate along the southwest coast of Sumatra and Java， as revealed by the eastward moving positive SLAs from day（-10） to day（+5） （Figure 3（c）-（f））. It is also found that positive SLAs leap over the Sunda Strait on day（-10） and then arrive at Lombok on day（-5）. The ensuing propagation is divided into three pathways: propagating eastward throughout the Sumba Strait and Savu Sea to reach the Ombai Strait （Pathway A）； penetrating into Lombok and propagating northward to reach the Makassar Strait （Pathway B）； and propagating southeastward along the southwest coast of the Sumba Island （Pathway C）. Ittakes about fve days for the downwelling Kelvin waves to propagate from Lombok to Makassar and Ombai （Figure 3（d） and （e））. The positive SLA lasts for another fve days and diminishes in the following days （Figure 3（f））. Meanwhile，there are easterlies over the equatorial Indian Ocean around day（0）， which drive upwelling Kelvin waves that propagate eastward following the same waveguides （Figure 3（e）-（i））. Throughout the propagation of downwelling Kelvin waves，there are westerly or onshore wind along the Sumatra and Java coast， favoring the enhancement of the Kelvin waves，and vice versa. This is corroborated by the increasing SLA during its propagation.

Figure 3.Composite intraseasonal SLAs and sea surface wind anomalies over the equatorial Indian Ocean and Indonesian seas.

Figure 4.Lag correlations of intraseasonal SLAs between diferent straits or passages of the ITF region and equatorial western Sumatra coast.

The pathways are further confrmed by the lag correlations between the key areas and/or straits and passages in the ITF region （Figure 4）. Figure 4（a）-（c） show the lag correlations along Pathway A. The results show that the maximum correlation between Sunda and Sumatra is 0.81，with Sunda lagging by 2 days； the maximum correlation between Lombok and Sumatra is 0.64， with Lombok lagging by 6 days； and the maximum correlation between Ombai and Sumatra is 0.35， with Ombai lagging by 12 days. The lag correlation analysis shows that the Kelvin waves take 2 days to travel for more than 540 km from Sumatra to Sunda， suggesting a phase speed of about 3.15 m s-1. Similarly， the phase speeds for the Kelvin waves from Sunda to Lombok and from Lombok to Ombai are estimated as 2.86 and 1.91 m s-1-in agreement with that of the frst and second baroclinic mode of the theoretical Kelvin wave， respectively.

The lag correlation analysis for Pathway B shows a maximum correlation of 0.31/0.32 between the north of Lombok Strait in the Java Sea/Makassar Strait and Sumatra， with the former lagging by six and nine days，respectively （Figure 4（d）-（f））. This suggests that the phase speed for downwelling Kelvin waves propagating from Lombok to Makassar is around 1.69 m s-1-in agreement with that of the second baroclinic Kelvin waves. The correlation between Sumba and Sumatra is 0.27 at the time lag of nine days， which is three days later than Lombok Strait， and suggests a phase speed of 1.96 m s-1along Pathway C. The lag correlations between Timor and Sumatra are much smaller， and thus barely provide useful information for the propagation from Sumba to Timor.

The Hovmöller diagrams shown in Figure 5 indicate the eastward propagation of SLAs from the central equatorial Indian Ocean to the Ombai and Makassar straits and the Timor Passage along Pathways A， B， and C. The equatorial Kelvin waves take about 14 days to travel from the equatorial Indian Ocean （70°E） to the western coast of Sumatra，which are trapped by the eastern boundary of the Indian Ocean. The ensuing coastal Kelvin waves then propagate along the southwest coast of the Lesser Sunda Islands. After arriving at Lombok Strait， the Kelvin waves divide into three branches， with more energy intruding into the Makassar Strait than through the Sumba Strait to reach the Ombai Strait， evidenced by the fact that there are larger SLAs in the Makassar than Ombai Strait. The other branchalong Pathway C shows an eastward propagation terminating at Savu Strait， which is reminiscent of the lack of correlation in SLAs between the Timor Passage and equatorial western Sumatra.

Figure 5.Hovmöller diagrams of the 20-90-day band-passed SLAs along the diferent pathways of intraseasonal Kelvin waves.

4. Summary

Gridded SLA data provided by AVISO are used in this study to track the propagation of intraseasonal Kelvin waves in the ITF region. The large RMS of intraseasonal SLAs along the Sumatra and Java coast is closely related to the propagation of intraseasonal Kelvin waves that derive from the equatorial Indian Ocean. These Kelvin waves are further found to propagate along diferent pathways in the Lombok Strait （Figure 1）. Through lag correlation analysis and Hovmöller diagrams of SLAs， we estimate the phase speed of the intraseasonal Kelvin waves for Pathways A， B，and C to be 1.91-2.86， 1.69， and 1.96 m s-1， respectivelyin agreement with the frst two baroclinic modes of Kelvin waves.

Disclosure statement

No potential confict of interest was reported by the authors.

Funding

This work was jointly supported by the National Natural Science Foundation of China （NSFC） ［grant numbers 41476025，41506036， 41306031］； NSFC-Shandong Joint Fund for Marine Science Research Centers ［grant number U1406404］； China Postdoctoral Science Foundation Funded Project ［grant number 2014M561883］； Postdoctoral Innovation Foundation of Shandong Province ［grant number 201403019］.

References

Arief， D.， and S. P. Murray. 1996. “Low-frequency Fluctuations in the Indonesian Throughfow through Lombok Strait.” Journal of Geophysical Research： Oceans 101: 12455-12464.

Broecker， W. 1991. “The Great Ocean Conveyor.” Oceanography 4: 79-89.

Chong， J. C.， J. Sprintall， S. Hautala， W. L. Morawitz， N. A. Bray，W. Pandoe. 2000. “Shallow Throughfow Variability in the Outfow Straits of Indonesia.” Geophysical Research Letters 27（1）: 125-128.

Dee， D. P.， S. M. Uppala， A. J. Simmons， P. Berrisford， P. Poli， S. Kobayashi， U. Andrae， et al. 2011. “The ERA-Interim Reanalysis: Confguration and Performance of the Data Assimilation System.” Quarterly Journal of the Royal Meteorological Society 137: 553-597.

Drushka， K.， J. Sprintall， S. T. Gille， and I. Brodjonegoro. 2010.“Vertical Structure of Kelvin Waves in the Indonesian Throughfow Exit Passages.” Journal of Physical Oceanography 40: 1965-1987.

Feng， M.， and S. E. Wijfels. 2002. “Intraseasonal Variability in the South Equatorial Current of the East Indian Ocean.” Journal of Physical Oceanography 32 （1）: 265-277.

Ffeld， A.， and A. L. Gordon. 1996. “Tidal Mixing Signatures in the Indonesian Seas.” Journal of Physical Oceanography 26: 1924-1937.

Gordon， A. L. 1986. “Interocean Exchange of Thermocline Water.”Journal of Geophysical Research 91 （4）: 5037-5046.

Gordon， A. L.， J. Sprintall， H. M. Van Aken， R. D. Susanto， S. E. Wijfels， R. Molcard， A. Ffeld， W. Pranowo， and S. Wirasantosa. 2010. “The Indonesian Throughfow during 2004-2006 as Observed by the INSTANT Program.” Dynamics of Atmospheres and Oceans 50: 115-128.

Iskandar， I.， W. Mardiansyah， Y. Masumoto， and T. Yamagata. 2005. “Intraseasonal Kelvin Waves along the Southern Coast of Sumatra and Java.” Journal of Geophysical Research 119: C04013. doi:http://dx.doi.org/10.1029/2004JC002508.

Iskandar， I.， Y. Masumoto， K. Mizuno， H. Sasaki， A. K. Afandi， D. Setiabudidaya， and F. Syamsuddin. 2014.“Coherent Intraseasonal Oceanic Variations in the Eastern Equatorial Indian Ocean and in the Lombok and Ombai Straits from Observations and a Highresolution OGCM.” Journal of Geophysical Research 119: 615-630. Molcard， R.， M. Fieux， and A. G. Ilahude. 1996. “The Indo-Pacifc Throughfow in the Timor Passage.” Journal of Geophysical Research： Oceans 101: 12411-12420.

Pujiana， K.， A. L. Gordon， J. Sprintall， and R. D. Susanto. 2009.“Intraseasonal Variability in the Makassar Strait Thermocline.”Journal of Marine Research 67: 757-777.

Pujiana， K.， A. L. Gordon， and J. Sprintall. 2013. “Intraseasonal Kelvin Wave in Makassar Strait.” Journal of Geophysical Research 118: 2023-2034.

Qiu， B.， M. Mao， and Y. Kashino. 1999. “Intraseasonal Variability in the Indo-Pacifc Throughfow and the Regions Surrounding the Indonesian Seas.” Journal of Physical Oceanography 29: 1599-1618.

Schiller， A.， S. E. Wijfels， J. Sprintall， R. Molcard， and P. R. Oke. 2010. “Pathways of Intraseasonal Variability in the Indonesian Throughfow Region.” Dynamics of Atmospheres and Oceans 50: 174-200.

Sprintall， J.， S Wijfels， A. L. Gordon， et al. 2004. “INSTANT: A New International Array to Measure the Indonesian Throughfow.”EOS Transactions American Geophysical Union 85 （39）: 369-376.

Sprintall， J.， S. E. Wijffels， R. Molcard， and I. Jaya. 2009.“Direct Estimates of the Indonesian Throughflow Entering the Indian Ocean: 2004-2006.” Journal of Geophysical Research 114: C07001. doi:http://dx.doi.org/10.1029/ 2008JC005257.

Syamsudin， F.， A. Kaneko， and D. B. Haidvogel. 2004. “Numerical and Observational Estimates of Indian Ocean Kelvin Wave Intrusion into Lombok Strait.” Geophysical Research Letters 31: L24307. doi:http://dx.doi.org/10.1029/2004GL021227.

Wijfels， S. E.， and G. Meyers. 2004. “An Intersection of Oceanic Waveguides: Variability in the Indonesian Throughfow Region.” Journal of Physical Oceanography 34: 1232-1253.

Yang， G.， W. D. Yu， Y. L. Yuan， X. Zhao， F. Wang， G. X. Chen， L. Liu，and Y. L. Duan. 2015. “Characteristics， Vertical Structures， and Heat/Salt Transport of Mesoscale Eddies in the Southeastern Tropical Indian Ocean.” Journal of Geophysical Research 120: 6733-6750. doi:http://dx.doi.org/10.1002/2015JC011130.

Yuan， D. L.， J. Wang， T. F. Xu， P. Xu， H. Zhou， X. Zhao， Y. H. Luan，W. P. Zheng， and Y. Q. Yu. 2011. “Forcing of the Indian Ocean Dipole on the Interannual Variations of the Tropical Pacifc Ocean: Roles of the Indonesian Throughfow.” Journal of Climate 24: 3593-3608.

Yuan， D. L.， H. Zhou， and X. Zhao. 2013. “Interannual Climate Variability over the Tropical Pacifc Ocean Induced by the Indian Ocean Dipole through the Indonesian Throughfow.”Journal of Climate 26: 2845-2861.

组

印度尼西亚贯穿流； Kelvin波； 季节内变化； 波导； 传播路径

15 March 2016

CONTACT WEI Ze-Xun weizx@fo.org.cn

© 2016 The Author（s）. Published by Informa UK Limited， trading as Taylor & Francis Group.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License （http://creativecommons.org/licenses/by/4.0/）， which permits unrestricted use， distribution， and reproduction in any medium， provided the original work is properly cited.