or collective redistirbutionor collective of this by of any article portion Th THE INDONESIAN published in been has is article Oceanography

SEA SURFACE , Volume 4, a quarterly journal of Th 18, Number photocopy machine, reposting, or other means reposting, photocopy machine, is only w permitted TEMPERATURE AND ITS VARIABILITY IN THE INDONESIAN Society. Copyright e Oceanography 2005 by Th ith the approval of Th

BY TANGDONG QU, YAN DU, JANE STRACHAN, GARY MEYERS, AND JULIA SLINGO e Oceanography Society. All rights reserved. Permission reserved. Society. All rights is e Oceanography gran

Lying at the confl uence of the Eurasian sequently, regional dynamics and lationship to SST fascinating to study. all correspondence to: Society. Send e Oceanography [email protected] or Th Plate, the Indo-Australian Plate, and SST are important factors in regional The SSTs across the Indonesian region the Pacifi c Plate is the Indonesian ar- climate, with important consequences are of paramount importance to convec- chipelago. It is composed of more than for global climate. tion and, hence, precipitation distribu- 3,000 islands, covering a global surface The Indonesian region, also known tion (Figure 1b). Warm SSTs around the area equivalent to the continental United as the “Maritime ,” has been islands of the Maritime Continent lead States and with a cumulative coastal pe- identifi ed as an area of major climatic to vast amounts of evaporation trigger- rimeter that is more than twice ’s importance both locally and globally. ing surface heat fl ux, which drive the circumference. The mechanisms that The region, along with equatorial Af- deep convective cell over the region. The generate and maintain surface tem- rica and , is recognized as relationship between SST and convec- ted in teaching to copy thisand research. for use Repu article perature (SST) (Figure 1a) within the a primary energy source for the entire tive activity is highly sensitive; modeling Indonesian seas are a consequence of the global circulation system. The main sup- studies (Miller et al., 1992) and observa- e Oceanography Society, PO Box 1931, Rockville, MD 20849-1931, USA. 1931, Rockville, Box PO Society, e Oceanography complex topography and connectivity ply of energy is latent heating, released tions (McBride et al., 2003) show that between the Pacifi c and Indian . from the condensation of water vapor small changes in the SST within the Mar- In addition to surface heat fl uxes, intense when clouds and precipitation form due itime Continent can result in signifi cant tidal mixing of surface and thermocline to cumulus convection. Convective ac- changes in precipitation patterns across waters and variability in thermocline tivity, which dominates weather in the the Indo-Pacifi c region (Ashok et al., depth driven remotely by winds over the tropics, is variable on a range of spatial 2001; Neale and Slingo, 2003). The effect Pacifi c and Indian Oceans play a role in and temporal scales, making the climate of these changes on the economy and en- generating and maintaining SST. Con- of the Maritime Continent and its re- vironment of the region is widespread. blication, systemmatic reproduction,

50 Oceanography Vol. 18, No. 4, Dec. 2005 Figure 1. Tropical Rainfall Measuring Mission’s (TRMM) (Kummerow et al., 1998) (a) sea surface temperature (°C) and (b) precipitation (103 mm/yr) averaged from December 1997 to December 2003. Warm sea surface tempera- ture in the region leads to strong convection and signifi cant precipitation.

Oceanography Vol. 18, No. 4, Dec. 2005 51 The Maritime Continent, with its huge providing a new view of SST fi elds and peak-to-peak amplitude exceeds 4°C range of variability in atmospheric and other ocean-atmosphere variables in (Figure 2a). The amplitude is much larg- oceanic activity, is a major modeling the Indonesian region at unprecedented er than in the surrounding heat pools challenge. Many sophisticated atmo- resolutions in both time and space. This in both the western Pacifi c and eastern spheric General Circulation Models paper provides a brief overview of the , where the amplitude (GCMs) do not agree with observa- recent studies that have identifi ed pro- is less than 1.5°C. On the interannual tions. With improving data on regional cesses controlling SST variability and time scale (Figure 2b), the highest SST variability (> 4.0°C) occurs along and offshore the and coasts, indicative of a strong remote infl uence of The Indonesian region SST is of major the equatorial Indian Ocean combined importance to atmospheric state, not only with local upwelling (discussed later). The interannual variability in the Timor, over the region itself, but globally. Convection, Arafura, and Banda Seas is smaller (less which is dependent on the SST, is the dominant than 2.0°C). atmospheric process over the region. INFLUENCE OF THE MONSOON ON SST The Asian monsoon has a dominant precipitation, and model validation, we related interactions with the atmosphere. infl uence on SST variation. In August, recognize that rainfall is systematically We begin with a description of general when the southeast monsoon prevails underestimated. Neale and Slingo (2003) characteristics of SST variability, present (Figure 3a), a broad area south of 5°S argue that the defi cient rainfall over the data on the sensitivity of the atmosphere cools, with the temperature minima in Maritime Continent could be respon- to underlying SST, and then continue the upwelling zone south of Java and sible for systematic errors elsewhere, with a survey of local and remote oce- over the Arafura Shelf. The cool waters both in the tropics and extratropics. anic mechanisms that affect SST. Finally, appear to be carried into the eastern Java The discrepancy in rainfall suggests in- we discuss the issues that need to be ad- adequate representation of the physical dressed by further research. Tangdong Qu ([email protected]) is system. Having a good understanding Associate Researcher, International Pacifi c of observed SST variability as it relates GENERAL CHARACTERISTICS Research Center, School of Ocean and Earth to regional ocean dynamics is key to im- OF SST VARIABILITY Science and Technology, University of Ha- proved model simulations. SST variability in the region is gener- waii, Honolulu, HI, USA. Yan Du is Postdoc- SST variability is also an important ally small compared with that in the toral Fellow, International Pacifi c Research infl uence on ’s marine ecologi- tropical eastern Pacifi c due to the lack of Center, School of Ocean and Earth Science cal systems, which harbor more than 20 strong equatorial upwelling. However, and Technology, University of Hawaii, percent of the world’s species of plants as with the surrounding western Pacifi c Honolulu, HI, USA. Jane Strachan is Ph.D. and animals. The inhabitants of the and eastern Indian Ocean warm pools, student, NCAS-Centre for Global Atmo- region, or at least a signifi cant part of the mean SST is high (Figure 1a), and spheric Modelling, University of Reading, UK. them, depend on these marine resources relatively small variations play an impor- Gary Meyers is Senior Principal Research for food, and many for their livelihood. tant role in coupled ocean-atmosphere Scientist, CSIRO Marine Research, Hobart, The rapid advance in space-based mi- processes. The largest seasonal SST cycle Tasmania, . Julia Slingo is Director, crowave remote sensing is revolution- occurs in the Timor, Arafura, Banda, and NCAS-Centre for Global Atmospheric Mod- izing ocean observations, in particular the South China Seas, where the SST elling, University of Reading, UK.

52 Oceanography Vol. 18, No. 4, Dec. 2005 Sea, then fl ow northwestward and enter accurate forcing of convective processes, and Slingo, 2003). The land-sea surface the through Karimata particularly on the diurnal time scale. temperature contrasts, along with the Strait. To compensate the outfl ow in the The complex system of islands, nar- coastal and orographical detail, are cru- surface layer, relatively cold water is draw row peninsulas, and shallow seas give cial for this complex diurnal cycle, one of from below, which apparently contrib- rise to extensive land-sea breeze circu- the most important modes of convective utes to the cooling in the southern Java lations and gravity wave effects (Yang variability in the region. However, global and Flores Seas (Wyrtki, 1961). In the and Slingo, 2001) that can infl uence the atmospheric models currently misrepre- Strait, where Coriolis param- energy and moisture budgets and hence sent the diurnal cycle over the islands in eter is close to zero, surface water fl ows the mean climate of the region (Neale both phase and amplitude. This shortfall northward, primarily in the direction of the wind. The effect of the surface fl ow is reduced by the enhanced subsurface fl ow from the Pacifi c (Susanto and Gor- don, 2005), and as a result, the SST in the is still higher than 29°C during this season. The monsoon wind reversal drives the surface layer of the Java Sea south- eastward (Figure 3b), bringing relatively cool, fresh South China Sea water into the region (Wyrtki, 1961). Within the internal Indonesian seas, the SST is low- est in the western Java Sea, while the enhanced surface heat fl ux makes SST higher than 28°C in the rest of the archi- pelago. The South Java Current (Quadfa- sel and Cresswell, 1992; Bray et al., 1996) 1.5 2 2.5 3 3.5 4 4.5 5 5.5 warms the area south of Java in this sea- son, as discussed in more detail later.

SENSITIVITY OF THE ATMOSPHERE TO UNDERLYING SST The Tropical Rainfall Measuring Mis- sion (TRMM) Microwave Imager (TMI) has provided a description of the SST patterns in Indonesia in much greater spatial detail than ever before. The coarse resolution used by global models poorly represents the geographical detail of the Maritime Continent. At low resolution, Figure 2. Peak-to-peak amplitude of (a) seasonal and (b) interannual variability in TRMM sea surface temperature (°C) from December 1997 to June 2004. Th e largest seasonal cycle (> 4.0°C) the high SSTs in the shallow coastal re- occurs in the Timor, Arafura, Banda, and South China Seas. Th e largest interannual variability gions are lost. This detail is crucial to the (> 4.0°C) occurs along and off shore the Java and Sumatra coasts.

Oceanography Vol. 18, No. 4, Dec. 2005 53 may be due to a combination of inaccu- comparison Project, second phase] is SST just within Indonesia (Figure 4b) racies in the geographical representation, a standard experimental protocol for results in a signifi cant change in precipi- prescribed SSTs, and parameterization of global atmospheric GCMs) (Gates, 1992) tation distribution in a direction that be- convective processes. shows systematic differences throughout gins to correct the model dry bias. Fur- Climate-model experiments that the region as large as 1°C. Experiments ther, applying this regional SST anomaly show the sensitivity of the atmosphere run on the atmosphere-only version of also has remote impacts across the whole to Indonesian SST are just beginning the UK Meteorological Offi ce Unifi ed Indo-Pacifi c region. A 1°C anomaly to emerge (Figure 4a). A careful com- Model (HadGAM1) show that using the applied over Indonesia results in a sig- parison (not presented) of TMI-SST to AMIP2-SST produces a dry bias in rain- nifi cant decrease in precipitation over the often-used coarse-grid AMIP2-SST fall over the Maritime Continent (Figure the western Pacifi c and western Indian (AMIP2 [Atmospheric Model Inter- 4a), as mentioned earlier. A 1°C rise in Oceans, areas which without the forcing

Figure 3. Monthly mean TRMM sea surface temperature (°C) superimposed with Quick Scatterometer (QuikSCAT) wind (m/s) in (a) August and (b) Febru- ary. Data are averaged from December 1997 to June 2004 and from July 1999 to January 2005, respectively. Here, the monsoon infl uence is evident. Th e sea surface temperature tends to warm up in the downstream direction of the wind and cool down in the upstream direction.

54 Oceanography Vol. 18, No. 4, Dec. 2005 show a wet bias (Figure 4a). Applying es up to 6 mm/day compared to increas- review some of the oceanic mechanisms similar SST anomalies elsewhere in the es up to 18 mm/day with a 1°C anomaly. that ultimately will have to be taken into Indo-Pacifi c region does not produce re- Additionally, the 0.5°C anomaly produc- account in regional climate modeling. mote responses. These results reveal the es much weaker remote effects. signifi cance of SST in the Maritime Con- These results show that the atmo- TIDAL MIXING AND tinent to the whole of the tropics. sphere over the Maritime Continent is UPWELLING A similar experiment with a 0.5°C highly sensitive to changes in SST. It is The Indonesian region has the stron- anomaly (Figure 4c) shows that the con- clear that the SSTs must be accurately gest tidal currents of the tropical ocean vective response over the region does prescribed in models in order to correct- (Robertson and Ffi eld, this issue). These not appear to be linear. Applying a 0.5°C ly reproduce the observed atmospheric currents generate intense ocean mix- anomaly produces precipitation increas- response. In the following sections, we ing, which infl uences SST in numerous

(a) control-CMAP observational data 30N Figure 4. Precipitation re- 20N sults from the atmosphere- only version of the UKMO 10N UM (UK Meteorological EQ Offi ce Unifi ed Model),

10S HadGAM1 (Hadley Centre climate model). (a) Th e 20S error in the model when 30S compared against pre- 0 60E 120E 180 120W 60W cipitation data from the Climate Prediction Centre’s (b) +1K Maritime Continent SST anomaly-control (CPC) Merged Analysis

30N of Precipitation (CMAP) (Xie and Arkin, 1996). 20N (b) Th e change from the 10N control model when a 1°C

EQ anomaly is applied to the Maritime Continent region 10S (90°E–160°E, 17.5°S–15°N).

20S (c) Th e change from the control model when a 30S 0 60E 120E 180 120W 60W 0.5°C anomaly is applied to the Maritime Continent region. Th e results reveal (c) +0.5K Maritime Continent SST anomaly-control the signifi cance of sea sur- 30N face temperature in the 20N Maritime Continent to the precipitation of the whole 10N tropics. EQ

10S

20S

30S 0 60E 120E 180 120W 60W -12 -6 -2 2 6 10 14 20 mm/day -8 -4 0 4 8 12 16

Oceanography Vol. 18, No. 4, Dec. 2005 55 channels and basins of the archipelago (Figure 5) shows high values of chlo- correlation between ITF strength, depth (Ffi eld and Robertson, this issue). Tidal- rophyll concentration in the upwelling of thermocline, and SST. Seasonally, the mixing-induced changes in SST modu- areas as well as within 200–500 km of the maximum impact of ITF on SST usually late ocean-atmosphere heat exchange coast off most Indonesian islands, imply- comes in late northern summer when as well as convective rainfall. Recent ing that strong tidal mixing and seasonal the pressure gradient between the Pacifi c studies have found that the strong fort- upwelling also infl uence the nutrient and the Indian Ocean is greatest (Wyrtki, nightly component of tides correlates concentrations, which further infl uence 1987; Meyers et al., 1995). with a fortnightly change in rainfall over fi shery and biological productivity in the Another important pathway is water the Indonesian archipelago: strong tides region (Hendiarti et al., this issue). entering the Java Sea from the South lead to a cooler ocean surface and re- China Sea through (Qu duced rainfall (e.g., Ffi eld and Gordon, REMOTE INFLUENCE FROM et al., 2004). On the seasonal time scale, 1996). Other links of tides to regional THE PACIFIC this fl ow has a maximum impact on SST climate are being investigated. The Indonesian throughfl ow (ITF) in boreal winter when the Luzon Strait The seasonally reversing monsoon (Gordon, 1986; Godfrey, 1996; Lee et Transport is strongest and the bifurcation also creates upwelling. Wyrtki (1962) al., 2002; Gordon, this issue) is infl uen- of the North Equatorial Current (NEC) fi rst showed the location of upwelling tial in regulating regional SST. Without occurs at its northernmost position (Qu in the region, based on early in situ - the constant supply of warm ITF water and Lukas, 2003). In response, the buoy- servations. The TMI (Wentz et al., 2000) and a deep thermocline transmitted ant, low-salinity South China Sea sur- provides a detailed picture of seasonal from the western equatorial Pacifi c, the face water then fl ows into the southern upwelling in the region (Figure 3). The Indonesian region, as well as the entire Makassar Strait, creating a northward southern coasts of Java and Flores Island, southeastern tropical Indian Ocean, pressure gradient in the surface layer of the southern coast of Irian Jaya, and the would be as dry and cold as the west the strait (Gordon et al., 2003), inhibiting eastern in particular are areas coast of South America (Wajsowicz and the ITF’s warming effect. of extensive upwelling. Schneider, 2001). In the Makassar Strait, SST variability at a broad range of The impact of mixing and upwelling for example, recent mooring observa- time scales is a consequence of the extends beyond SST. Satellite ocean color tions (Ffi eld et al., 2000) showed the unique position of Indonesia at the con-

Figure 5. Average July to September chlorophyll

concentration (log10 mg/ m3) 1997–2003, from Sea- viewing Wide Field-of-view Sensor (SeaWiFs). Th e high chlorophyll concentration during this season off ers evidence for the exis- tence of upwelling within 200–500 km of the coast, despite its indistinct sig- nature in the sea surface temperature.

56 Oceanography Vol. 18, No. 4, Dec. 2005 fl uence of the Pacifi c and Indian equato- Gordon, 2005) are well captured by the mocline (Potemra, 2001; Wijffels and rial and coastal waveguides (Clarke and satellite’s TMI measurements (Figure Meyers, 2004). Variations in zonal winds Liu, 1994). Thus, the depth of the ther- 6a). Moreover, Ffi eld et al. (2000) found in the central Indian Ocean generate mocline and SST in the region can be that the Makassar thermocline tempera- equatorial Kelvin waves, which propa- infl uenced by remote equatorial winds ture is highly correlated (r=0.77) to the gate eastward along the equatorial wave- in the Pacifi c and Indian Oceans. This Southern Oscillation Index (SOI). guide. Upon approaching the Indonesian mechanism allows the El Niño-Southern coast, the refl ected equatorial Kelvin Oscillation (ENSO) to infl uence regional REMOTE INFLUENCE FROM waves push the thermocline down along SST (Meyers, 1996; Potemra, 2001; Wi- THE INDIAN OCEAN the Sumatra-Java-Nusa coast, chang- jffels and Meyers, 2004). The cold SST SST in the region is also infl uenced by ing the mixed layer depth and hence the anomalies in Makassar Strait during the variations in wind over the equatorial response to surface fl uxes or upwelling peak of 1997/98 El Niño (Susanto and Indian Ocean, transmitted in the ther- and mixing. A strong manifestation of

Figure 6. (a) Anomalous TRMM sea surface temper- ature (°C) superimposed with Special Sensor Micro- wave Imager (SSM/I) wind (m/s) and (b) anomalous TRMM precipitation (103 mm/mon) in December 1997. Th e cold sea surface temperature and negative precipitation anomalies near the coast of Indonesia were well captured, indica- tive of the occurrence of an Indian Ocean Dipole event (Saji et al., 1999).

Oceanography Vol. 18, No. 4, Dec. 2005 57 this process, occurring twice a year dur- over the (Anna- chlorophyll-concentration water perme- ing monsoon transitions, is known as the malai and Slingo, 2001). A positive IOD ates from the coast to about 98°E. But, Wyrtki Jets (Wyrtki, 1973). event is usually accompanied by heavy why does the upwelling not appear in the Recently, as a result of the occurrence rainfall in eastern and droughts in pattern of SST, as it does in other eastern of extreme events in 1994 and 1997, con- Indonesia (Figure 6b). boundary ? A detailed answer to this question illustrates many of the dif- fi culties that must be faced in developing ...it is important to accurately prescribe SSTs an adequate simulation of regional SST in a climate model. in the Maritime Continent in climate models This question was fi rst raised by Wyrt- so that they will correctly reproduce the ki (1962), but has received little attention response in the atmosphere, and hence better until recent studies of Quadfasel and Cresswell (1992), Qu et al. (1994), and predict future climatic conditions. Du et al. (2005). The small SST depres- sion off Java and Sumatra results from internal and external factors. The region siderable attention has been focused on SEASONAL UPWELLING OFF is unique, partly because of the existence an ocean-atmosphere coupled phenom- JAVA AND SUMATRA of the ITF. The ITF transfers a large enon known as the Indian Ocean Dipole SST variability off Java and Sumatra is a amount of warm, fresh water from the (IOD) mode (Saji et al., 1999), which is key component of IOD and is strongly Pacifi c to the Indian Ocean, and affects also known as the Indian Ocean Zonal coupled to the seasonal cycle. In the both the regional circulation and ther- (IOZ) mode (Webster et al., 1999). The usual upwelling season (June to August), mal structure (Qu and Meyers, 2005). A positive phase of IOD is characterized we do not see a SST depression of any quantitative analysis of the upper-layer by an east-west dipole pattern in SST more than a few tenths of a degree cen- heat budget (Qu et al., 1994), based on anomalies spanning the Indian Ocean tigrade off Java and Sumatra (Figure 3a), results from a coarse-resolution (0.5° basin. Like the Pacifi c ENSO, the evolu- despite the upwelling-favorable south- x 0.5°) GCM, indicates that seasonal tion of IOD is strongly locked to the an- east monsoon. This is highly unusual warming advection by the ITF cancels nual cycle. In its active years, cold SST compared to other upwelling regions off the cooling by upwelling near the coast anomalies off Java and Sumatra typically North and South America and Africa. of Java. This cancellation effect explains develop in June to August and peak in The high chlorophyll concentration at why upwelling doesn’t appear in the pat- September to October (Figure 6a), while this season offers evidence for the exis- tern of SST. The emergence of higher- warm SST anomalies in the western In- tence of upwelling within 200–500 km of resolution (≤ 0.1°) GCMs allows a fresh dian Ocean occur later. The SST anoma- the coast (Figure 5). As wind blows from look at this problem. lies are found to be closely associated the southeast, water is drawn from below Despite some differences in quanti- with changes in surface winds. Equato- and carried offshore in the surface layer, tative detail, a recent study (Du et al., rial winds reverse direction from west- inducing widespread upwelling along 2005) using results from a high-resolu- erlies to easterlies during the peak phase the coast of Indonesia (Wyrtki, 1962; tion GCM confi rms Qu et al.’s (1994) of IOD when SST is cold in the east and Susanto et al., 2001). The upward move- interpretation for the small SST depres- warm in the west (Figure 6a). Changes in ment brings high-nutrient water from sion south of Java. In addition to the ITF, surface winds are associated with a ba- the thermocline into the euphotic layer Du et al. (2005) found that the barrier sin-wide anomalous Walker circulation where phytoplankton develop under layer, representing an intermediate layer (Yamagata et al., 2002), as well as anom- increasing light penetration. As part of that separates the base of the mixed layer alous local Hadley cell and monsoons the southwestward Ekman current, high- from the top of the thermocline (Lukas

58 Oceanography Vol. 18, No. 4, Dec. 2005 and Lindstrom, 1991), is another unique Sumatra is controlled primarily by the A number of recent ocean GCM experi- factor infl uencing the SST variability in winds over the equatorial Indian Ocean. ments show that the barrier-layer thick- the region. To the west of Sumatra, in The Pacifi c and Indian Ocean wind fi elds ness, rainfall, and SST anomalies are particular, large rainfall and runoff per- are not tightly coupled and they display tightly correlated in the region near Su- sist year round. The enhanced surface substantially different time scales of matra on interannual time scales (Mur- stratifi cation sustains a shallow mixed variability (Wijffels and Meyers, 2004). tugudde and Busalacchi, 1999; Masson layer, but a thick barrier layer (Sprintall and Tomczak, 1992; Masson et al., 2002; Qu and Meyers, in press). From boreal summer to fall, the mixed layer is rela- tively shallow. Because of the topograph- ically oriented rainfall and runoff, con- tours of mixed-layer depth tend to fol- low the coastline all the way from south Java to west Sumatra (Figure 7a). The southwestward Ekman current forced by southeast monsoon creates a cross- shelf pressure gradient that generates the northwestward South Java Current (SJC) (Quadfasel and Cresswell, 1992). The SJC advects low-salinity surface water stemming from the large P-E (precipita- tion minus evaporation) (~10 mm/day) to the equator and horizontally homog- enizes the salinity distribution along the coast (e.g., Masson et al., 2002). Coastal upwelling, however, is shallow (< 50 m) and has little effect on the deeper ther- mocline. The presence of a thick barrier layer (Figure 7b) impedes cold thermo- cline water from entering the mixed lay- er, and this explains why the depression of SST off Sumatra is small compared with other eastern boundary upwelling regions like that west of Peru in the east- ern Pacifi c (Du et al., 2005). Analysis of the seasonal heat budget sets the background for further study of interannual variability in the region off Java and Sumatra. As noted above, Figure 7. (a) Monthly mean mixed layer depth and (b) barrier layer thickness (m) from July to ITF variability is primarily controlled by September (adapted from Qu and Meyers, in press). Th e existence of a thick (> 20 m) barrier the winds over the Pacifi c Ocean, while layer explains why the sea surface temperature depression off Java and Sumatra is small com- pared with other eastern boundary upwelling regions like that west of Peru in the eastern Pa- depth of the thermocline off Java and cifi c (Du et al., 2005).

Oceanography Vol. 18, No. 4, Dec. 2005 59 et al., 2004). In terms of the correlation cycle. These processes, however, become not always be maintained on interannual between SST and thermocline depth less important on longer time scales; time scales. This occasional imbalance anomalies, this region is important for the interannual SST variability can be leads to substantially different realiza- the growth of IOD in the Indian Ocean. largely understood in terms of Kelvin tions of the seasonal heat budget, thus al- and Rossby waves generated by remote lowing for strong upwelling signature in DISCUSSION zonal winds along the Indian and Pacifi c the sea surface in some particular events. The Indonesian region SST is of major equators. The volume transport of ITF No observational evidence is available importance to atmospheric state, not is another important factor infl uencing to support a heat-budget analysis of this only over the region itself, but globally. water temperature in the region. This speculation. The promise of recent high- Convection, which is dependent on the infl uence is primarily in the depths of resolution GCM results warrants a more SST, is the dominant atmospheric pro- the thermocline, but it infl uences SST in extensive exploration to determine fac- cess over the region. Convective activity regions of upwelling and strong vertical tors that initiate the cold SST anomalies drives atmospheric circulation and con- (tidal) mixing. off Java/Sumatra, as well as modulate tributes to global energy and moisture The directly monsoon-driven circula- their evolution in the IOD events. budgets. Therefore, it is important to ac- tion in the surface layer of Java Sea con- Finally, we note that the Indonesian curately prescribe SSTs in the Maritime tributes to the seasonal ITF variability as region has suffered, and will continue Continent in climate models so that they well as SST in Makassar Strait. This result to suffer, from scarcity of data and in will correctly reproduce the response in implies an important oceanic connection particular, of regionally consistent data. the atmosphere, and hence better predict between the tropical Pacifi c Ocean and Although weather-station-based SST future climatic conditions. the Indonesian seas. As a source and sink observations from Indonesia and nearby We have provided a general descrip- of the Java Sea’s surface fl ow, the South countries have been analyzed individu- tion of the SST variability in the Indo- China Sea is believed to play an impor- ally, there have been few attempts to nesian region. As the SST variability in tant role in the connection. On inter- collect these data sets and quantify the the region has a great impact on regional annual time scales, the infl ow of cold, spatial and temporal variability of SST climate, environment, economy, and relatively fresh South China Sea water over the entire region. Recent satellite biological productivity, we hope such a through Karimata Strait has its maxi- observations provide for the fi rst time description will be useful for the design mum strength around the mature phase an overall view of SST variability in the and analysis of future observations. In of El Niño (Qu et al., 2004). A natural region, but the characteristics have not addition, we have surveyed recent stud- question is how this infl ow contributes to been systematically compared with those ies to address the questions of whether the ITF and regional SST. This question derived from the much longer and lo- and how oceanic processes play a role in needs to be addressed by future studies. cally dense weather-station-based data generating the observed SST anomalies The largest SST variability in the Indo- sets. Retrieving data from historical ar- of the region. nesian region occurs in the coastal region chives in Indonesia and nearby countries The SST distribution in the region is off Java and Sumatra. Despite the prevail- is probably the most cost-effective way complex, varying considerably in both ing southeasterly winds in boreal sum- of obtaining long-term SST observations time and space. This complexity makes mer, the upwelling signature is present in in the region. it diffi cult to understand the processes the SST pattern only during IOD events. that control the SST variability. Differ- In normal years, the SST depression ACKNOWLEDGEMENTS ent processes are dominant on different across the Java/Sumatra coast is small, The National Aeronautics and Space time scales in different parts of the re- presumably as a result of warming advec- Administration supported this research gion. Within the Indonesian archipelago, tion by ITF and the formation of a thick through grant NAG5-12756. The Japan local monsoon and tidal mixing are barrier layer. We expect that the delicate Marine Science and Technology Center important, and probably explain a sig- balances of advection, upwelling/entrain- also supported TQ through its sponsor- nifi cant part of the mean seasonal SST ment, and barrier-layer formation will ship of the International Pacifi c Research

60 Oceanography Vol. 18, No. 4, Dec. 2005 Center (IPRC), and GM by the CSIRO Mission (TRMM) sensor package. Journal of At- Ocean dynamics in the region between Australia Wealth from Oceans Flagship Program. mospheric and Oceanic Technology 15:809–817. and Indonesia and its infl uence on the variation Lee, T., I. Fukumori, D. Menemenlis, D. Zhang, and of sea surface temperature in a global general The authors are grateful to A.L. Gor- L. Fu. 2002. Effects of the Indonesian through- circulation model. Journal of Geophysical Research don for his kind invitation to make this fl ow on the Pacifi c and Indian Oceans. Journal of 99:18,433–18,445. Physical Oceanography 32:1,404–1,429. Quadfasel, D., and. G.R. Cresswell. 1992. A note on contribution, and to E.J. Lindstrom and Lukas, R., and E. Lindstrom. 1991. The mixed layer the seasonal variability of the South Java Current. an anonymous reviewer for valuable of the western equatorial Pacifi c Ocean. Journal of Journal of Geophysical Research 97:3,685–3,688. comments on the earlier version of this Geophysical Research 96(Suppl.):3,343–3,357. Saji, N.H., B.N. Goswami, P.N. Vinayachandran, and Masson, S., J.P. Boulanger, C. Menkes, P. Delecluse, T. Yamagata. 1999. A dipole mode in the tropical manuscript. IPRC contribution num- and T. Yamagata. 2004. Impact of salinity on the Indian Ocean. Nature 401:360–363. ber IPRC-348, and School of Ocean and 1997 Indian Ocean dipole event in a numerical Sprintall, J., and M. Tomczak. 1992. Evidence of the experiment. Journal of Geophysical Research 109: barrier layer in the surface layer of the tropics. Earth Science and Technology contribu- C02002, doi: 10.1029/2003JC001807. Journal of Geophysical Research 97:7,305–7,316. tion number 6655. Masson, S., P. Delecluse, J.P. Boulanger, and C. Menk- Susanto, R.D., and A.L. Gordon. 2005. Velocity and es. 2002. A model study of the seasonal variability transport of the Makassar Strait throughfl ow. and formation mechanisms of the barrier layer Journal of Geophysical Research 110:C01005, REFERENCES in the eastern equatorial Indian Ocean. Journal doi:10.1029/2004JC002425. Annamalai, H., and J.M. Slingo. 2001. Active/Break of Geophysical Research 107:8017, doi:10.1029/ Susanto, R.D., A.L. Gordon, and Q. Zheng. 2001. Up- Cycles: Diagnosis of the intraseasonal variability 2001JC000832. welling along the coasts of Java and Sumatra and of the Asian summer monsoon. Climate Dynam- McBride, J.L., M.R. Haylock, and N. Nichols. 2003. its relation to ENSO. Geophysical Research Letters ics 18:85–102. Relationships between the maritime continent 28:1,599–1,602. Ashok, K., Z. Guan, and T. Yamagata. 2001. Impact of heat source and the El Niño–Southern Oscillation Wajsowicz, R.C., and E.K. Schneider. 2001. The In- the Indian Ocean Dipole on the relationship be- phenomenon. Journal of Climate 16:2,905–2,914. donesian throughfl ow’s effect on global climate tween Indian Ocean monsoon rainfall and ENSO. Meyers, G. 1996. Variation of Indonesian through- determined from the COLA coupled climate Geophysical Research Letters 28:4,499–4,502. fl ow and the El Niño-Southern Oscillation. Jour- system. Journal of Climate 14:3,029–3,042. Bray, N.A., S. Hautala, J. Chong, and J. Pariwono. nal of Geophysical Research 101:12,255–12,263. Webster, P.J., A.M. More, J.P. Loschnigg, and R.R. 1996. Large-scale sea level, thermocline, and wind Meyers, G., R.J. Bailey, and A.P. Worby. 1995. Geo- Leban. 1999. Coupled ocean-atmosphere dynam- variations in the Indonesian throughfl ow region. strophic transport of Indonesian throughfl ow. ics in the Indian Ocean during 1997–98. Nature Journal of Geophysical Research 101:12,239– Deep-Sea Research Part I 42:1,163–1,174. 401:356–360. 12,254. Miller, M.J., A.C.M. Beljaars, and T.N. Palmer. 1992. Wentz, F.J., C. Gentemann, D. Smith, and. D. Chelton. Clarke, A.J., and X. Liu. 1994. Interannual sea level in The sensitivity of the ECMWF model to the pa- 2000. Satellite measurements of sea surface tem- the northern and eastern Indian Ocean. Journal rameterization of evaporation from the tropical perature through clouds. Science 288:847–850. of Physical Oceanography 24:1,224–1,235. oceans. Journal of Climate 5:418–434. Wijffels, S., and G. Meyers. 2004. An intersection of Du, Y., T. Qu, G. Meyers, Y. Masumoto, and H. Sasa- Murtugudde, R., and A.J. Busalacchi. 1999. Interan- oceanic waveguides: variability in the Indonesian ki. 2005. Seasonal heat budget in the mixed layer nual variability of the dynamics and thermody- throughfl ow region. Journal of Physical Oceanog- of the southeastern tropical Indian Ocean in a namics of the tropical Indian Ocean. Journal of raphy 34:1,232–1,253. high-resolution global general circulation model. Climate 12:2,300–2,326. Wyrtki, K. 1987. Indonesian throughfl ow and the as- Journal of Geophysical Research 110:C04012, Neale, R.B., and J.M. Slingo. 2003. The Maritime sociated pressure gradient. Journal of Geophysical doi:10.1029/2004JC002845. Continent and its role in the global climate: A Research 92:12,941–12,946. Ffi eld, A, and A.L. Gordon. 1996. Tidal mixing signa- GCM study. Journal of Climate 16:834–848. Wyrtki, K. 1973. An equatorial jet in the Indian tures in the Indonesian Seas. Journal of Physical Potemra, J.T. 2001. Contribution of equatorial Pa- Ocean. Science 181:262–264. Oceanography 26:1,924–1,937. cifi c winds to southern tropical Indian Ocean Wyrtki, K. 1962. The upwelling in the region be- Ffi eld, A., K. Vranes, A.L. Gordon, and D. Susanto. Rossby waves. Journal of Geophysical Research tween Java and Australia during the south-east 2000. Temperature variability within Makassar 106:2,407–2,422. monsoon. Australian Journal of Marine and Strait. Geophysical Research Letters 27:237–240. Qu, T., and G. Meyers. In press. Seasonal variation Freshwater Research 13:217–225. Gates, W.L. 1992. AMIP: The Atmospheric Model of the barrier layer in the southeastern tropical Wyrtki, K. 1961. Physical oceanography of Southeast Intercomparison Project. Bulletin of the American Indian Ocean. Journal of Geophysical Research. Asian waters. Naga Report 2. Scripps Institution Meteorological Society 73:1,962–1,970. Qu, T., and G. Meyers. 2005. Seasonal characteris- of Oceanography, La Jolla, CA, 195 pp. Godfrey, J.S. 1996. The effect of the Indonesian tics of circulation in the southeastern tropical Xie, P.P., and P.A. Arkin. 1996. Analyses of global throughfl ow on ocean circulation and heat ex- Indian Ocean. Journal of Physical Oceanography monthly precipitation using gauge observations, change with the atmosphere: A review. Journal of 35:255–267. satellite estimates and numerical model predic- Geophysical Research 101:12,217–12,237. Qu, T., and R. Lukas. 2003. On the bifurcation of the tions. Journal of Climate 9:840–858. Gordon, A.L. 1986. Interocean exchange of ther- North Equatorial Current in the Pacifi c. Journal Yamagata, T., S.K. Behera, S.A. Rao, Z. Guan, K. mocline water. Journal of Geophysical Research of Physical Oceanography 33:5–18. Ashok, and H.N. Saji. 2002. The Indian Ocean 91:5,037–5,046. Qu, T., Y.Y. Kim, M. Yaremchuk, T. Tozuka, A. dipole: a physical entity. CLIVAR Exchange Gordon, A.L., R.D. Susanto, and K. Vranes. 2003. Ishida, and T. Yamagata. 2004. Can Luzon Strait 24:15–18, 20–22. Cool Indonesian throughfl ow as a consequence of Transport play a role in conveying the impact of Yang, G.Y., and J.M. Slingo. 2001. The diurnal cycle in restricted surface layer fl ow. Nature 425:824–828. ENSO to the South China Sea? Journal of Climate the tropics. Monthly Weather Review 129:784–801. Kummerow, C., W. Barnes, T. Kozu, J. Shiue, and J. 17:3,644–3,657. Simpson. 1998. The Tropical Rainfall Measuring Qu, T., G. Meyers, J.S. Godfrey, and D. Hu. 1994.

Oceanography Vol. 18, No. 4, Dec. 2005 61