Makassar Strait is the primary pathway of the
Pacific to Indian Ocean transport referred to as the Indonesian
throughflow. The transport through Makassar Strait was measured
as part of the Indonesian-USA Arlindo program, at two moorings
deployed within the Labani Channel, a deep (2000 m) constriction
(45 km) near 3°S (Fig. 1). Both moorings
were operative from December 1996 to February 1998, a 1.4 year
time series, when the MAK-2 mooring was released and recovered;
the MAK-1 mooring was recovered in early July providing a 1.7
year record. The MAK moorings were deployed during a weak La Niña
phase. An El Niño condition began in March 1997, becoming
extreme during 1997 summer and fall, relaxing in early 1998. Arlindo
data will be the subject of much study by the Arlindo research
team**, but because of WOCE interest in Indonesian Throughflow
we offer this preview of Makassar transport based on the Aanderaa
current meters (current, temperature, pressure) and temperature-pressure
pods of MAK-1 and MAK-2.
The Indonesian maritime continent with its complex network of passages and basins connecting the Pacific and Indian Oceans inhibits free communication between the Pacific and Indian Oceans. Schneider (1998) using a couple model, shows that the presence of the Pacific to Indian interocean transfer shifts the warmest SST and associated atmospheric convective region towards the west, relative to a no throughflow condition. Webster et al. (1998) state: the Indonesian throughflow heat flux "...is comparable to the net surface flux over the northern Indian Ocean and a substantial fraction of the heat flux into the western Pacific warm pool...it would appear that the throughflow is an integral part of the heat balances of both the Pacific and Indian Oceans."
Observations indicate that the throughflow is composed mostly of North Pacific thermocline and intermediate water flowing through Makassar Strait (Gordon and Fine, 1996), which then passes into the Indian Ocean through the passages of Lesser Sunda Islands. East of Sulawesi South Pacific water infiltrates the lower thermocline and dominates the deeper layers, including the Lifamatola Passage overflow into the deep Banda Sea (Van Aken et al., 1988; Gordon and Fine, 1996; Hautala et al. 1996), but it is unlikely that the eastern channels carry total more than 3 Sv.
Indonesian throughflow estimates based on observations, models and conjecture range from near zero to 30 Sv (Godfrey, 1996). Measurements in the Lombok Strait in 1985 (Murray and Arief, 1988; a near zero SOI value) indicate an average transport of 1.7 Sv. Molcard et al. (1996) as part of the French-Indonesian program JADE, find a mean transport to the Indian Ocean of 4.3 Sv between the sea surface and 1250 m in the Timor Passage from March 1992 to April 1993 (an El Niño period). With the Lombok values the JADE results suggest 6 Sv transport through the Lesser Sunda Islands. An annual mean throughflow of 5 Sv is estimated from XBT data for the upper 400 meters between Java and Australia for the period 1983 to 1989 (Meyers, 1996). T.H. Aung presented the results of the 1993-94 (El Niño period) ASEAN current meter array in the Makassar Strait at a June 1995 meeting in Lombok. The three ASEAN moorings were deployed in the wide northern entrance to Makassar Strait. Most of the current meters were below 400 m with one instrument at 275 m, thus missing the main thermocline, making estimation of transport difficult, but Aung states that the Makassar transport may be as large as 11 Sv. Potemra et al (1997) model study and inspection of TOPEX/POSEIDON data find a summer maximum of 11 Sv, and a winter minimum of 4 Sv, with a 7.4 Sv 9-year mean. Gordon et al. (1997) find on average 9 Sv of Indonesian throughflow water advected westward within the Indian Ocean.
The preliminary findings of the Arlindo Makassar MAK-1 and MAK-2 data are presented in Figs 2 , 3 and 4, some key points each of which will be explored in detail by the Arlindo team, are:
1. The Makassar thermocline depth and transport reflect the phases of ENSO, with an ambiguous seasonal cycle: deeper thermocline, larger throughflow during La Niña; shoal thermocline, with reduced transport during El Niño. Additionally, during the El Niño months December 1997 to February 1998 the transport average is 5 Sv, while during the La Niña months of December 1996 to February 1997 the average is 12 Sv, a 2.5 fold difference.
2. Along channel flow exhibits much activity at frequencies above seasonal. A special event occurs in May and June 1997 when a marked relaxation of the throughflow transport is recorded. Candidates responsible are: Pacific Ocean Rossby waves, Indian Ocean coastal Kelvin waves, local atmosphere and dynamics internal to the Indonesian Seas.
3. The Makassar Strait 1997 twelve month average throughflow is 9.3 Sv. This assumes that the flow above the shallowest Aanderaa equals the flow at that current meter (case B, Fig. 4). Other models for the surface flow yield 1997 transport average of 6.7 Sv (zero surface flow, case C, Fig. 4) to 11.3 Sv (thermocline shear is extrapolated to the sea surface, case A, Fig. 4). How to handle the water flow above the shallowest Aanderaa current meter is an important issue, not just for the mass transport but also for the interocean heat and freshwater flux and for monitoring array design. We will have a firmer idea of the surface layer flow when the moored ADCP data are processed. The MAK-2 ADCP has a record from 1 December 1996 to 9 March 1997 before it flooded; the MAK-1 ADCP data record will be processed later this year. The preliminary MAK-2 ADCP data show a maximum of along channel flow at 110 m, with near zero surface flow. The hull ADCP of the Baruna Jaya IV, the Indonesian research vessel used in the Nov/Dec 1996 deployment and Feb 1998 recovery of the MAK moorings reveal similar reduction of along channel speed in the surface layer. The MAK-1 monthly along channel speeds displays higher southward values at the 250 m instrument relative to the 200 m instrument for 13 of the 20 month record (the months with higher transport). The data suggests shear reversal between 200 and 250 m in the western Labani channel, closer to 100 m in the east.
4. The Makassar transport (case B) determined from the Arlindo data is at the higher end of estimates derived from Timor Sea and Indian Ocean studies. While this would favor the case C throughflow of 6.7 Sv, with zero mean along channel flow at the sea surface this may not be the only explanation. Perhaps we are seeing a throughflow interannual (ENSO) signal (noting that the JADE Timor Passage values were obtained during an El Niño period)? Alternatively, might some of the Makassar transport pass back to the Pacific Ocean to the east of Sulawesi? Comparison of the MAK mooring results with: 1996-98 JADE mooring data near Timor (Molcard and Fieux); Lesser Sunda Island shallow pressure gauge array (Janet Sprintall); and data from the Arlindo mooring in Lifamatola Passage (Fig. 1) to be recovered in November 1998, may help resolve this issue.
Acknowledgment: The research is funded by NSF: OCE 95-29648 and the Office of Naval Research: N00014-98-1-0270. Gratitude and appreciation is extended to Dr. Indroyono Soesilo and to Basri M. Ganie of BPPT for arranging for the joint program and the ship time aboard Baruna Jaya IV, to Gani Ilahude of LIPI for all that he has done to promote the Arlindo project, and to Captain Handoko and his fine staff aboard the Baruna Jaya IV. And finally we thank Baldeo Singh of UNOCAL for the recovery of MAK-1 mooring in July 1998.
* on leave from The Agency for Assessment and
Application of Technology (BPPT) Jakarta, Indonesia.
** The US scientists involved in Arlindo Circulation are: Arnold L. Gordon (US chief scientist for Arlindo; CTD) and Amy Ffield (temperature-pressure pods), Lamont-Doherty Earth Observatory of Columbia University; Silvia Garzoli (PIES), NOAA Atlantic Oceanography and Meteorology Laboratory; Dale Pillsbury (current meters) Oregon State University; Rana Fine (CFC) University of Miami. Chester Koblinsky (satellite data), Goddard Space Flight Center. The Arlindo Indonesian Team include: R. Dwi Susanto (BPPT and LDEO); A.G. Ilahude (LIPI and LDEO); from BPPT: Muhamad Irfan, Handoko Manoto, Fadli Syamsudin, Djoko Hartoyo, Bambang Herunadi; and from LIPI: Muhamad Hasanudin.