Air-Sea Interaction, Coastal Circulation and Primary ...svr4.terrapub.co.jp/journals/JO/pdf/6002/60020205.pdfduces upwelling which lowers sea surface temperature by about 2.5°C (compared

205

Journal of Oceanography, Vol. 60, pp. 205 to 218, 2004

Review

Keywords:⋅ Air-sea interaction,⋅ monsoons,⋅ coastal circulation,⋅ water masses,⋅ upwelling,⋅ primary produc-tion,⋅ chlorophyll-a,⋅ eastern ArabianSea.

* Corresponding author. E-mail: [email protected] address: National Center for Antarctic & Ocean Research,

Govt. of India, Headland Sada, Vasco-da-Gama, Goa 403804, India.

Copyright © The Oceanographic Society of Japan.

Air-Sea Interaction, Coastal Circulation and PrimaryProduction in the Eastern Arabian Sea: A Review

ALVARINHO J. LUIS* and HIROSHI KAWAMURA

Center for Atmospheric and Oceanic Studies, Faculty of Science, Tohoku University,Sendai 980-8578, Japan

(Received 14 November 2002; in revised form 21 April 2003; accepted 19 May 2003)

Air-sea interaction, coastal circulation and primary production exhibit an annualcycle in the eastern Arabian Sea (AS). During June to September, strong southwest-erly winds (4~9 m s–1) promote sea surface cooling through surface heat loss andvertical mixing in the central AS and force the West India Coastal Currentequatorward. Positive wind stress curl induced by the Findlater jet facilitates Ekmanpumping in the northern AS, and equatorward-directed alongshore wind stress in-duces upwelling which lowers sea surface temperature by about 2.5°C (compared tothe offshore value) along the southwestern shelf of India and enhances phytoplanktonconcentration by more than 70% as compared to that in the central AS. During win-ter monsoon, from November to March, dry and weak northeasterly winds (2–6m s–1) from the Indo-China continent enhance convective cooling of the upper oceanand deepen the mixed layer by more than 80 m, thereby increasing the vertical flux ofnutrients in the photic layer which promotes wintertime phytoplankton blooms in thenorthern AS. The primary production rate integrated for photic layer and surfacechlorophyll-a estimated from the Coastal Zone Color Scanner, both averaged for theentire western India shelf, increases from winter to summer monsoon from 24 to 70g C m–2month and from 9 to 24 mg m–2, respectively. Remotely-forced coastal Kelvinwaves from the Bay of Bengal propagate into the coastal AS, which modulate circula-tion pattern along the western India shelf; these Kelvin waves in turn radiate Rossbywaves which reverse the circulation in the Lakshadweep Sea semiannually. This re-view leads us to the conclusion that seasonal monsoon forcing and remotely forcedwaves modulate the circulation and primary production in the eastern AS.

ing an intense low-level jet (Findlater, 1977) over thecentral AS. In response to these winds a clockwise circu-lation evolves in the AS (Wyrtki, 1971; Schott, 1983;Swallow, 1984). The equatorward eastern boundary of thisanticyclonic circulation is known as the West IndiaCoastal Current (WICC) (Shetye et al., 1990). To the southof Sri Lanka, the WICC merges with the eastward flow-ing Southwest Monsoon Current (SMC) which bendsaround the Sri Lankan coast and flows poleward into theBay of Bengal (Fig. 1(a)).

During the winter or northeast monsoon, generallyfrom November to February, the winds blow from thenortheast. These winds force the Northeast MonsoonCurrent (NMC) toward the west (Fig. 1(b)). Part of thisflow bifurcates at the southwest Indian coast and flowspoleward to form the WICC (Fig. 1(b)) (Shetye et al.,

1. IntroductionOne of the highest primary productions among the

world oceans occurs in the Arabian Sea (AS) due to avariety of physical processes introduced by the semian-nual reversal of monsoon winds (Qasim, 1982; Banse,1987). This basin, which forms the western flank of thenorth Indian Ocean, offers a striking example of wind-driven ocean circulation (Wyrtki, 1973). During the south-west or summer monsoon, generally from June throughSeptember, strong winds blow from the southwest form-

206 A. J. Luis and H. Kawamura

1991). This poleward flow, which occurs in opposition tothe wind field, is facilitated by a density gradient alongthe west coast of India (WCI) (Shetye and Shenoi, 1988).Reduced solar radiation and high evaporative cooling inthe northern AS lead to high salinity water (>36 ppt), whilelow salinity water occupies the region to the south of theIndian tip (IT). This establishes a density gradient whichleads to the poleward flow of the WICC. During theintermonsoon period, from March to April and Octoberto November, weak, highly variable wind regimes (2~3m s–1) occur in the AS (Hastenrath and Lamb, 1979a) andthe basin surface circulation dissipates (Cutler andSwallow, 1984). The southwest monsoon period is char-acterized by sea surface cooling in the central AS(Colborn, 1976; Düing and Leetmaa, 1980) due to sur-face heat loss and entrainment (Hastenrath and Lamb,1979b; Shetye, 1986; Bauer et al., 1991). While a posi-tive wind stress curl to the north of the Findlater jet pro-motes Ekman pumping during the summer monsoon, con-vective cooling during the winter monsoon leads todensification and downwelling in the northern AS.

In this review we highlight the studies carried outalong the eastern AS, extending from the Pakistan coast(25.5°N) to south of Sri Lanka (5.5°N) (dashed line in

Fig. 1(a)), and identify “gray” areas that need further at-tention. The purpose of emphasizing on only the easternAS is that recent statistics indicate that of the annual fishcatch, which ranges from 2.2 to 2.8 million tonnes (Cen-tral Marine Fisheries Research Institute, 1995), about 73%of the catch originates from the west coast of India due toa high primary production (PP) in this sector of the AS.The continental shelf, which we identify by the 500-mcontour in Fig. 1(a), is narrow (50 km) south of Karachi,widens to ~350 km off the Gulf of Cambay and graduallytapers towards south to ~120 km at the south Indian tip(IT).

At this point, it is relevant to note that the dynamicprocesses in the eastern AS are induced by local and re-mote forcing. During the summer monsoon, offshore di-vergence of the alongshore wind stress component leadsto coastal upwelling and sea surface temperature (SST)cooling (Shetye et al., 1985; Muraleedharan and Kumar,1996; Naidu et al., 1999; Luis and Kawamura, 2002a, b).Strong summer monsoon winds enhance evaporative cool-ing in the central AS (McCreary and Kundu, 1989). Onthe other hand, wind jets in the equatorial Indian Ocean,between 5°S to 5°N, excite equatorial Kelvin waveswhich, on reflection from the eastern boundary of the Bay

Fig. 1. (a) Geography of the northern Arabian Sea. This review focuses on the region identified by a thick dashed line. Schemat-ics of summer-monsoon circulation are superimposed. Ekman pumping region in the northern Arabian Sea is highlighted inyellow tone. Coastal upwelling promoted by divergence of alongshore wind stress component is indicated in green tone.Current branches indicated are the Ras al Hadd Jet (RHJ), Lakshadweep Low (LL), West India Coastal Current (WICC),Southwest Monsoon Current (SMC), Sri Lanka Dome (SD) and East India Coastal Current (EICC). The boxes (a), (b), (c) and(d) are referred in Figs. 7 and 10. The Findlater Jet and wind direction are indicated by bold gray arrows. (b) As in (a), but forwinter monsoon. Convective cooling region is shown in yellow tone. Additional abbreviations shown are: Lakshadweep High(LH) and Northeast Monsoon Current (NMC).

Air-Sea Interaction, Coastal Circulation and PP in the Eastern Arabian Sea: A Review 207

of Bengal, propagate along the perimeter of this basin asa coastal Kelvin wave and radiate Rossby waves (Yu etal., 1991). The coastal Kelvin waves propagate along theperimeter of the Bay of Bengal, bend around the SriLankan coast and propagate poleward along the WCI. Thedownwelling (upwelling) Kelvin wave radiatesdownwelling (upwelling) Rossby waves which propagateoffshore and promote anticyclonic (cyclonic) circulationin the Lakshadweep Sea during the winter (summer)monsoon (Bruce et al., 1994; Shankar and Shetye, 1997;Bruce et al., 1998). The anticyclonic Lakshadweep cir-culation is also strengthened by negative wind stress curlduring the winter monsoon.

In Section 2 we describe the monsoon-ocean inter-action. Section 3 is devoted to coastal circulation, whilePP and chlorophyll-a (Ch-a) mapping studies are high-lighted in Section 4. The relationship between the physi-cal dynamics and PP is also explored in Section 3. A dis-cussion and summary are offered in Section 4.

2. Monsoon-Ocean InteractionDuring the northern hemisphere summer, the solar

heating of the Indian subcontinent cause a low pressurearea over Arabia and Pakistan, establishes a north-southpressure gradient of about 22 hPa (Tomczak and Godfrey,1994) and drives the southwest monsoon. On the otherhand, wintertime cooling of the landmass causes a south-ward pressure gradient of about 6 hPa that drives the win-ter monsoon. The summer monsoon is maintained bypropagation of convective systems of synoptic (lows,depressions, etc.) and planetary scale (tropical conver-gence zones, TCZ) from warm tropical oceans onto theheated subcontinent (Gadgil, 2000). The TCZ is charac-

terized by cyclonic vorticity associated with large shearof the zonal wind in the north-south direction, at all lev-els in the lower troposphere above the atmospheric bound-ary. In this section we focus on seasonal forcing on theeastern AS using ship-based measurements, satellite-based observations and modeling studies.

2.1 Ship-based evidenceFigure 2 depicts the fields of monsoon-ocean inter-

action parameters and surface heat budget, which wereestimated using 60 years of ship observations (Hastenrathand Lamb, 1979a, b). These are presented at 1° × 1° spa-tial resolution. The top (bottom) row represents the con-ditions during July (January) corresponding to the sum-mer (winter) monsoon. During July the wind speed is high(8~9 m s–1) along the central WCI and westerly windswith 4–6 m s–1 speed occur south of 12°N (Fig. 2(a)).The winter monsoon is characterized by northerly tonortheasterly winds with reduced strength compared tothat of the summer monsoon (Fig. 2(b)). Strong winds (6m s–1) in the Gulf of Mannar (GM) are attributed tochanneling of the winds through the sea-level channelbetween the elevated landmass of southern India and SriLanka. These winds have been referred to as gap winds,and their characteristic features have been highlighted andstudied elsewhere (Luis and Kawamura, 2000, 2001,2002a).

The surface atmospheric heat budget during Julyexhibits a heat loss (~40 W m–2) slightly north of 10°Nand an equal gain in the south (Fig. 2(c)). In January thenet heat flux is weak and is directed into the ocean to thesouth of Saurashtra (Fig. 2(d)). Heat loss in the GM is atthe expenditure of high latent heat (Hastenrath and Lamb,

Fig. 2. Spatial patterns of (a) and (b) wind speed (m/s), (c) and (d) surface net heat flux (W/m2), (e) and (f) wind stress curl(10–8 dynes/cm3), and (g) and (h) sea surface temperature (°C). The upper (lower) panel represents summer (winter) monsoon.The data resolution is 1° × 1°. Source: Hastenrath and Lamb (1979a, b).


1979b, Chart 45). The wind stress curl is favorable toupwelling along the entire coastal region during July, withhighly positive curl induced by the Findlater jet to westof Karachi (Fig. 2(e)). During January, a highly negativecurl (–3 × 10–8 dynes cm–3) to the southwest of the IT(Fig. 2(f)) facilitates an anticyclonic circulation in theLakshadweep Sea (Bruce et al., 1994, 1998). During July,positive wind stress curl and the divergence of thealongshore wind stress component promote upwelling andSST cooling along the southern WCI (Shetye, 1984). Thewinter monsoon SST scenario is as follows. The Ekmandivergence at the coast promotes upwelling; the offshoreadvection of this cold leads to low SST to the west of SriLanka (Luis and Kawamura, 2002a). Cold, low-salinitywater from the northern Bay of Bengal (BB) is alsoadvected into the eastern AS by the Northeast MonsoonCurrent (Fig. 1(b)), which lowers SST along the WCI(Haugen et al., 2002). Cold SSTs that occur in the north-ern region have been attributed to convective cooling inthe upper oceanic layers induced by dry continental winds(Banse, 1968).

Based on the analysis of Fig. 2 and with inferencesfrom the literature, the scenario during the monsoons isas follows. The southwest monsoon is more energetic thanthe winter monsoon. The magnitude of the offshore com-ponent of the wind during the summer monsoon is muchgreater than the alongshore component along the WCI(Shetye et al., 1985). With the alongshore wind stresscomponent directed toward the equator, upwelling occursalong the Kerala coast during both monsoons (Shetye,1984). During the winter monsoon, a high SST coolingincreases mixed layer thickness by ~100 m north of 20°N(Rao et al., 1989; Prasad, 1996) owing to a decrease insolar radiation and to high evaporation (Prasad, 1997)enhanced by dry northeasterly winds (Kumar and Prasad,1996).

2.2 Satellite-based evidenceSatellite-based monitoring offers data with repeated

periodicity, high spatial resolution and wide area cover-age, and provides an excellent view of the oceanographicand atmospheric features from the air-sea interfacial layerto the top of the cloud. The satellite oceanography startedwith the launch of SEASAT in 1978. Though this moni-toring lasted for only three months, it demonstrated thefeasibility of using satellites for geophysical studies. Afew satellite-based studies on the eastern AS have beencarried out and these provide useful insights into thephysical and biological forcing. Recently, Luis andKawamura (2000) employed daily-mean NASAScatterometer (NSCAT) winds and advanced very highresolution radiometer (AVHRR) 9-km spatial-resolutionSST and studied the role of wind on SST in the GM dur-ing January 1997. They argued that the sea-level channel

between the elevated land topography of Sri Lanka andsouth India facilitates channeling and acceleration of thenortheast trades through the GM (Fig. 3(a)); these windshave been referred as gap winds. The study indicated thatthe gap winds induce SST cooling of ~1°C due to deple-tion of surface heat through high latent heat loss (~220W m–2) (Fig. 3(a)).

Further analysis of the winds derived from the Spe-cial Sensor Microwave/Imager (SSM/I), AVHRR SSTsand meteorological variables from the National Centersfor Environmental Prediction/National Center for Atmos-pheric Research (NCEP/NCAR) reanalysis data set(Kalnay et al., 1996) indicated that wind stress and sur-face heat loss were punctuated in time ranging from 15to 30 days over the winter monsoon period. Furthermore,the turbulence caused by wind in the surface atmosphericboundary layer enhances the surface heat loss and lowersthe SST by 1.5°C over a 15-day time scale (Luis andKawamura, 2001).

2.3 Modeling studies of monsoon-ocean interactionNumerical modeling studies of the coastal air-sea

interaction in this sector of the AS are fragmentary. Shetye(1986) ran a simple mixed-layer model with Kraus-Turnerthermodynamics for a 2° latitude zonal strip (9–11°N),extending from Somalia to the Kerala coast, to understandevolution of the seasonal cycle of SST and mixed layer.The results indicated that momentum and heat forcing areeffective for the SST evolution during winter monsoonover the entire strip. In addition to surface forcing, theimportance of entrainment and advection to the evolu-

Fig. 3. Monsoon-ocean interaction monitored by satellites inthe vicinity of the Indian tip during (a) January 1997 and(b) May 1997. AVHRR sea surface temperature (gray tone)superimposed with contours of surface net heat flux(W m–2) and NASA Scatterometer (NSCAT) derived windstress vectors (N m–2). Source: Luis and Kawamura (2002a,b).


tion of coastal SSTs during summer monsoon is alsoemphasized.

For the first time, a high resolution (9 km), the three-dimensional, sigma-coordinate, primitive-equationPrinceton Ocean Model (POM) was run to understand thedynamics of the wintertime wind forcing in the GM (Luisand Kawamura, 2002a). The stratified POM was fittedwith realistic bottom topography and was forced withwinds observed by NASA Scatterometer (NSCAT) (Fig.3(a)) to gain insights into the evolution of SST during a14-day gap-wind event during 21 January to 3 February,1997. The model solutions revealed that coastal Ekmandynamics, together with positive wind stress curl, lowerSST west of Sri Lanka during the first 7 days of the event.The spatially variable and shallow shelf topography atthe IT promotes diapycnal mixing and alters the SST alongthe periphery of the IT. The mixed-layer heat budget re-vealed that the depletion of surface heat modulates SST.Based on the satellite-based observations (Luis andKawamura, 2000, 2001) and with inferences from themodel results, they concluded that SST cooling in the GMis related to the monsoon-sea-topography interaction.

Numerical study of SST cooling dynamics in the vi-cinity of the IT prior to the summer monsoon (May) (Fig.2(b)) also substantiated the role of physical forcing onthe ocean (Luis and Kawamura, 2002b). Using POM withrealistic bottom topography, and daily-mean NSCAT windforcing, these authors pointed out that wind-driven Ekmandynamics along the southwestern India coast facilitatesupwelling-induced SST cooling. An equatorward-directedcoastal jet advects the cold water southward, where theshallow, irregular shelf topography at the IT enhanceswind mixing, and a sharp gradient at the shelf break (~500m) modulates upwelling strength through cross-isobathflow from subsurface. They concluded that the southwardadvection of cold water generates low SSTs near the IT,in agreement with the AVHRR SST maps.

3. Coastal CirculationContrasting the eastern boundary current dynamics

of other oceans, the eastern AS show marked seasonalchanges in its eastern boundary currents. The coastal cir-culation in the eastern AS is fully three-dimensional. Weaddress this scenario from various dynamical viewpointsusing hydrographic and modeling approaches.

3.1 Hydrographic approach3.1.1 Water masses distribution

In the AS the seasonal evaporation minus precipita-tion (E–P) influences the surface salinity. During the sum-mer monsoon E–P lies between –1 and –1.5 m year–1 tothe east of 70°E and to the south of 20°N (Prasad, 1997);so high precipitation leads to the formation of EquatorialSurface Water (Darbyshire, 1967). The upper-ocean wa-

ter masses off the southwestern coast of India are charac-terized by Indian Equatorial Water (IEW) (Sverdrup etal., 1942; Emery and Meincke, 1986), with a contribu-tion from less saline Bay of Bengal Water (BBW) andmore saline Arabian Sea Water (ASW) (Fig. 4; Strammaet al., 1996). According to Darbyshire (1967), the north-ward WICC during the summer monsoon carries the lowdensity ESW to the north and ASW downdrafts and deep-ens the thermocline to about 100 m. During May to earlySeptember, ESW retreats and ASW shoals to depths lessthan 50 m and lowers SST (22°C) along the Kerala coast.

During the winter monsoon, E–P is positive and isnearly constant (1.5 m year–1) to the north of 10°N. Highevaporation caused by dry northerly winds enhances up-per ocean convection (Banse, 1968; Prasad, 1997) andincreases surface salinity, which leads to the formationof the Arabian Sea High Salinity Water mass (ASHSW)(Rochford, 1964; Kumar and Prasad, 1999). The charac-teristics of ASHSW are: temperature, 28–24°C; salinity,36.7–35.3 psu; and depth, 0–100 m. The ASHSW subductsin the region north of 20°N and spreads equatorward alongthe 24σt isopycnal. Below the surface is the water masscalled Red Sea Persian Gulf Intermediate Water (RSPGW;Emery and Meincke, 1986) or North Indian IntermediateWater (Wyrtki, 1973). This water mass is formed withthe contribution from Red Sea Water (RSW), which shows

Fig. 4. T-S distribution of conductivity-temperature-depth(CTD) measurements along 8°N and from 68.1°E to south-west shelf of India during 5–8 August, 1993. Stations to thewest (east) of 75°E are shown as dashed (solid) lines. Boxesindicate the range of temperature and salinity values forBay of Bengal Water (BBW), Arabian Sea Water (ASW),Indian Equatorial Water (IEW) and Red Sea Persian GulfIntermediate Water (RSPGW). Density surfaces of the coreof Persian Gulf Water (PGW) and Red Sea Water (RSW)are marked by arrows. Source: Stramma et al. (1996).


a maximum at about 300 m depth, and a contribution fromPersian Gulf Water (PGW), which shows a salinity maxi-mum between 550 to 800 m depth.

Figure 4 summarizes the temperature-salinity (T-S)characteristics of different water masses. It was con-structed from conductivity-temperature-depth (CTD)measurements along 8°N and spanning from 68°10′ E tothe southwest Indian coast during August 1993 (Strammaet al., 1996). The BBW is characterized by a T range of25 to 29°C and S of <35 psu at the surface, and is locatedbetween 74.2°E and the southern India shelf. During June–August 1987, BBW has been identified at about 150 mdepth with S below 35 psu and T between 14 to 20°C(Shetye et al., 1990). ASW mass is characterized by T >29°C and S > 35 psu. It is identified to the west of 75°E.IEW is identified to the east of 75°E and has a large rangeof T from 8 to 23°C and a close S range (34.5 to 35 psu).On the other hand, RSPGW is identified by low T (5–14°C) and S of about 35 psu. The density surfaces of thecore of PGW and RSW are indicated by arrows in Fig. 4.3.1.2 Upwelling

Banse (1968) summarized the upwelling events alongthe WCI as follows. (1) During the southwest monsooncool subsurface water occupies the entire shelf fromCochin (10°N) to Karachi (24.5°N); for the latter casecool water persists till November. (2) Low surface tem-peratures occur off Saurashtra and along the Pakistan coastdue to atmospheric cooling during winter monsoon.(3) Upwelling along the southwestern coast (south of10°N) during the southwest monsoon is due to seasonalchanges of mass distribution and to the anticyclonic cir-culation of the Arabian Sea, as suggested by Darbyshire(1967) and Wyrtki (1971).

During the UNDP/FAO Pelagic Fisheries Project(1971–75) about 200 hydrographic sections were occu-pied to the south of 17°N (Johannessen et al., 1981). Theynoted shoaling of isotherms at the southern coast fromApril to September. Besides the role of the local winds,they pointed out that the anticyclonic AS circulation con-tributes to coastal upwelling. However, Shetye and Shenoi(1988) strongly advocated that the wind-driven circula-tion dominates the WCI, as suggested by Wyrtki (1971).To further substantiate their claim, hydrocast data werecollected normal to the coast during the 1987 southwestmonsoon (Shetye et al., 1990) and with follow-up obser-vations during the 1987–88 northeast monsoon (Shetyeet al., 1991). During July–August they reported 2.5°Csurface cooling at the southern Indian coast, which wasattributed to upwelling promoted by the alongshore windstress component. The upwelling intensity increasednorthward from the IT as the alongshore wind compo-nent strengthened. As a signature of upwelling they iden-tified an undercurrent with its core at 150 m depth, whichwas discerned up to 20°N.

During the Indian Joint Global Ocean Flux Studies(JGOFS), oceanographic observations along 12.5, 15 and10°N were conducted during July–August, 1995 (Fig. 5)(Muraleedharan and Kumar, 1996). The section along12.5°N indicated signatures of strong upwelling, markedby cold SST (~22°C) and low salinity (35 psu) at the coast.As the upwelling intensity weakened, they inferred highsea surface salinity (>36 psu). At the 10°N section, theisotherms above 100 m exhibited shoaling at the coast.While all of the above studies reported upwelling due towind-driven Ekman dynamics, numerical solutions sug-gest that upwelling-favorable Kelvin waves from the Bayof Bengal propagate into the eastern AS and modulatecoastal circulation and SST along the WCI (McCreary etal., 1993; Haugen et al., 2002; Shankar et al., 2002).3.1.3 Coastal currents

Although a comprehensive hydrographic survey ofthe north Indian Ocean was conducted during the Inter-national Indian Ocean Expedition (IIOE) (Wyrtki, 1971),a detailed picture of the large-scale coastal circulationemerged only in the 1990s due to intensive oceanographic

Fig. 5. Vertical distribution of temperature (a, d, g), salinity(b, e, h) and density (c, f, i) along 15° (top panel), 12.5°(middle panel) and 15°N (bottom panel) during July–August, 1995. Unit: temperature, °C and salinity, psu. Re-produced from Muraleedharan and Kumar (1996).


surveys carried out by Indian scientists in the ExclusiveEconomic Zone (EEZ) of India. Considerable insightshave been gained through the analysis of hydrocast datacollected along sections normal to the WCI during the1987 southwest and 1987–88 northeast monsoons (Shetyeet al., 1990). During the summer monsoon they identi-fied an equatorward shallow (between 75–100 m) surfaceflow, with enhanced transport towards the south, and apoleward flowing undercurrent with its core at 150 m orbelow in close proximity to the continental slope.

The hydrographic data of December 1987 and Janu-ary 1988 provided insights into the poleward coastal cur-rent along the west coast in a situation where thealongshore wind component is directed towards the equa-tor (Fig. 1(b), Shetye et al., 1991). The presence of anundercurrent flowing equatorward was also inferred fromvertical density sections and dynamic topography. Basedon the alongshore momentum balance equation and withthe support from the inferences from the analytic modelof McCreary et al. (1986), who simulated the intensifica-tion of a poleward flowing Leeuwin Current and anequatorward undercurrent off western Australia, Shetyeet al. (1991) argued that the alongshore baroclinic pres-sure gradient was responsible for the maintenance of apoleward surface flow during the northeast monsoon.3.1.4 Coastal sea level

The sea level changes in the eastern AS are inducedby mass redistribution due to currents forced by monsoons(Figs. 1(a) and (b)) and by remote forcing from the Bayof Bengal. Locally, the seasonal sea level changes alongthe coast are attributed to the astronomical tides, atmos-pheric pressure, and thermohaline effects; the latter isdictated by heat and salt exchange across the oceanboundaries. Every wind system, whether stationary ormoving, will create currents, followed by redistributionof mass (Sverdrup et al., 1942). Coastal currents forcedby monsoons (Shetye et al., 1990; Shetye et al., 1991)also alter mass distribution through horizontal advection.Of these four factors, astronomical tides contribute lessthan 1 cm, which is an order of magnitude less than themeteorological effects (Shankar, 2000).

The climatological seasonal cycle of sea level (cor-rected for atmospheric pressure) exhibits an annual cy-cle: sea surface is elevated (depressed) during northeast(southwest) monsoon (Fig. 6). A good correspondencebetween sea level and alongshore currents inferred fromship drift at Marmagao (15°N), Mangalore (12.9°N), andCochin (10°N) is evident. However, a similar compari-son with the alongshore wind exhibited discrepancies,suggesting that remote forcing of the coastal circulationcould be important (Shankar, 2000). Shankar and Shetye(1999), using one-and-half layer and two-and-half layerreduced gravity models, attempted to relate salinity ef-fect to the sea-level variability along the coast. They noted

that lower (higher) salinity at the coast leads to higher(lower) coastal sea level. Since salinity changes are causedby precipitation and river runoffs, as pointed out by Sastryand D’Souza (1972), they inferred that the sea level alongthe coast can be linked to rainfall in particular, and themonsoon in general.

3.2 Modeling approachIn this section we highlight the remote forcing as-

pects on eastern AS circulation. A comprehensive numeri-cal investigation using a two-and-half layer thermody-namic model provided considerable insights into the lo-cally and remotely forced dynamics of the north IndianOcean (McCreary et al., 1993). These authors pointed outthat the winds along the eastern boundary of the Bay ofBengal set up coastal Kelvin waves, which bend aroundSri Lanka and propagate poleward along the WCI, forc-ing changes in the WICC (McCreary et al., 1993; Shankarand Shetye, 1997). The Kelvin waves originating in theBay of Bengal have their origins to the withdrawal andcollapse of the southwest monsoon during August–Octo-

Fig. 6. Climatological seasonal cycle of sea level anomalies(annual mean removed, solid line) and alongshore currents(dashed line) at seaports along the west India coast. Thesea level was corrected for atmospheric pressure. Redrawnfrom Shankar (2000).


ber. The role of remote forcing was further emphasizedby a test run without Arabian Sea winds; the results re-vealed that the coastal currents along the WCI were strong,but they were weaker without Bay-of-Bengal winds.

The remote forcing mechanism is also linked to theevolution of LH, which manifests as a large anticycloniceddy in the upper 300–400 m off the southwest India dur-ing January (Bruce et al., 1994) (Fig. 1(b)). The dynamictopographic maps indicated a dynamic topographic heightof about 15–20 cm at the eddy center, while SEASAT-based sea-surface height anomalies relative to the 1-yearmean (August 1987 to July 1988) showed a westwardmigration speed of 14–15 km day–1. Using a 3-layer, re-duced-gravity, wind-forced tracer model, they pointed outthat the Bay-of-Bengal low-salinity water is advected intothe core of the eddy via the Northeast Monsoon Current(NMC in Fig. 1(b)). This highlights the importance ofnegative wind-stress curl near the southwest coast of In-dia to the LH genesis. It was also noted that the Rossbywave radiations from downwelling Kelvin waves alongthe WCI also support the eddy genesis in late Decemberand its decay in April (Bruce et al., 1998).

Shankar and Shetye (1997), using an analytical modeland a one-and-half layer dynamically reduced-gravitymodel, pointed out that the sea level changes in theLakshadweep Sea exhibits an annual cycle, with a low(high) appearing during July (January). The results of themodel with an active layer of 100 m when forced withclimatological monthly-mean winds indicated that thedownwelling (upwelling) Rossby waves that are radiatedby downwelling (upwelling) and northward propagatingcoastal Kelvin waves generate the anticyclonic (cyclonic)eddy during January (July) (Figs. 1(b) and (a)). Theypointed out that only the Kelvin waves with period greaterthan 40 days were effective in the eddy formation and ininterannual and intraannual variability of the LH and LL.

In an attempt to address the mechanism of the for-mation of a warm SST pool in the Lakshadweep Sea priorto the onset of the summer monsoon (May), Shenoi et al.(1999) also invoked the remote forcing theory. Theypointed out that the downwelling Kelvin waves originat-ing from the western boundary of the Bay of Bengal forcean equatorward East Indian Coastal Current that carrieslow-saline Bay-of-Bengal water into the LakshadweepSea (Fig. 1(b)) and generates a stable surface layer, which,together with the downwelling Rossby waves radiatedfrom the IT, promote warm SSTs. This warm pool en-hances atmospheric moisture convergence and cause deepconvection, leading to the formation of the monsoon vor-tex during late May; this vortex forms a component ofthe monsoon system (Krishnamurthi et al., 1981).

4. Primary ProductionReversal in the surface circulation during monsoons

(Figs. 1(a) and (b)), seasonality in the nutrient distribu-tion (Banse, 1987) and irradiance (Brock et al., 1994) haveimportant ramifications for the PP in the north IndianOcean. In this section, we first delineate the seasonal dis-tribution of nutrients using climatology (World OceanAtlas-1998, 1999). Then we highlight phytoplankton dis-tribution using the ship borne measurements, model pre-dictions and satellite-based chlorophyll-a (Ch-a) obser-vations.

4.1 NutrientsFigure 7 displays seasonal vertical profiles of sili-

cate, nitrate, phosphate and oxygen at the locations iden-tified by box (a), (b), (c), and (d), respectively, in Fig.1(a). We compare the coastal ocean nutrient profiles withopen ocean profile sampled at 65.5°E, 12.5°N (panel (d)in Fig. 7). During the summer monsoon (top row) thevertical profiles exhibit the following features. At thenorthern shelf, the oxygen profile in the upper 150 m is

Fig. 7. Climatological profiles of nitrate, silicate, phosphateand oxygen during summer (top row) and winter monsoon(bottom row) at: (a) 23.5°N, 66.5°E, (b) 14.5°N, 73.5°E,(c) 6.5°N, 77.5°E, and (d) 12.5°N, 65.5°E (see boxes in Fig.1(a)). Data source: World Ocean Atlas-1998 (1999).


characterized by a sharp decline at a rate of 0.4ml l–1 m–1 in the upper 50 m and a maximum (20 ml l–1)at 125 m depth. Nitrate exhibits maxima at 100 and be-low 200 m. Silicate concentration increases below 125 mand the profile shows maxima at 200 m, 300 m and atdeeper depths. At the central location (box (b)), the sili-cate profile shows a maximum at 125 m, while nitrateconcentration increases above 20 µM as compared withthe northern box. The increase in the near-surface oxy-gen is the result of upwelling. For the southern box, ni-trate concentration exceeds 30 µM below 100 m and theoxycline exhibits a weaker gradient of 0.03 ml l–1 m–1.Compared to the coastal region, the open ocean profilesshow sharp contrasts. Higher vertical mixing in the cen-tral AS and offshore advection of upwelled water fromthe western AS (Kumar et al., 2000) enhances oxygenconcentration in the upper 75 m. The oxycline gradientincreases to 0.06 ml l–1 m–1.

The oxygen and nutrient profiles for winter monsoonare shown in the lower panel of Fig. 7. The followingfeatures are observed. At the northern box, strong con-vection and overturning in the surface layers increasesmixed layer depth, which enhances oxygen concentrationin the upper 140 m and renders a sharp oxycline (0.05ml l–1 m–1). Compared to the summer monsoon, the ni-trate profile exhibits a weak maximum at 200 m. For thecentral box, the oxygen concentration depletes in the up-per 150 m and nitrate concentration peaks at 200 m. Com-pared to the summer monsoon profile, the silicate con-centration exceeds 20 µM below 150 m. At the southernbox, higher oxygen concentration is evident in the upper50 m and nitrate concentration increases between 75 and300 m. The open-ocean profiles do not show any subsur-face maximum, but silicate concentration exceeds 30 µMbelow 300 m. In conclusion, the oxygen content of thesurface mixed layer is high and enhanced nutrient con-centration occurs just below the base of the surface mixedlayer during winter monsoon.

4.2 Primary production and phytoplankton bloomsHere the focus is on ship-based PP and phytoplankton

measurements and ecosystem models. Based on 21 cruisesin the northern AS north of 20°N, Qasim (1982) carriedout a comprehensive study of PP among other oceano-graphic aspects. He found that surface PP is high offMumbai, Gulf of Kutch and Saurashtra, and off Pakistan(Fig. 8). The annual average PP in coastal areas was re-ported to be ~37 mg C m–3, which was three times than inthe offshore regions. The values during the summermonsoon were one third of the value in other seasons.Madhupratap et al. (1996) advocated that the wintertimesea surface cooling drives convection processes that leadto the injection of nutrients into the surface waters, lead-ing in turn to high PP in the northeastern AS; the evi-

dence for this has been provided using Indian Ocean ColorMonitor (OCM) Ch-a composite images (Chauhan et al.,2001). As for the column PP (integrated over the euphoticlayer), Qasim’s computations indicated 270 × 106 tonnesC year–1, constituting more than 50% of the total produc-tion of the northern AS. The column-integrated PP peakedduring the summer monsoon. His study also showed thatabout 50% of the annual Ch-a production in the coastalsectors of the northeastern AS, constituting ~26 mg m–3

(15 × 103 tonnes year–1). The annual-mean surface PPbased on 14C uptake showed high values (50–100mg C m–3 day–1) along the central WCI and in the Gulf ofKutch, while high values of column-integrated PP overthe euphotic layer (1–2 g C m–2 day–1) are identified alongthe Saurashtra coast (Figs. 8(a) and (b)).

As a part of Indian JOGFS program, the seasonalityand composition of phytoplankton composition was de-ciphered along the WCI and compared with that alongthe 64°E meridian (Sawant and Madhupratap, 1996).These authors inferred higher phytoplankton populationdensity during the summer and winter monsoon period,which for the upper 120 m ranged from 1.7 to 170 and0.3 to 10 (×108) cells m–2, respectively. Their species-wise classification showed that diatoms constituted about86% and cyanobacteria and dinoflagellate were 7% and6% of the total population for all the seasons combined.Compared to the open ocean, the cell count (cells l–1) forthe coastal population of phytoplankton species waslarger. Among the diatom population Chaetoceros sp.topped the list with 75% higher counts in the coastal re-gion. This study suggested that physical mechanisms,

Fig. 8. Primary production in the northern Arabian Sea.(a) Annual mean rates of primary production (PP) based on14C uptake at the surface (mg C/m3/day) and (b) integratedPP for the euphotic zone (g C/m2/day). Gaps in the distri-butions indicate no data. Source: Qasim (1982).


among other constraints, could be attributed to higherphytoplankton counts in the coastal region. On the otherhand depletion in nutrients, especially nitrate, in the openocean waters restricts the PP (Gupta et al., 1980).

To isolate the physical mechanisms that promote thephytoplankton activity, McCreary et al. (1996) ran a four-component ecosystem model, with a 2 1/2 layer physicalmodel system with surface mixed layer embedded in theupper layer and a biological model with advective-diffu-sive equations in each layer that determine nitrogen con-centrations in four compartments: nutrients,phytoplankton, zooplankton and detritus. They comparedthe model solutions with climatological Coastal ZoneColor Scanner (CZCS) data. Their model solutions showupwelling blooms, detrainment blooms, and entrainmentblooms in the AS. The upwelling blooms are promotedby a high underwater irradiance and a large nutrient sup-ply to the shallow surface mixed layer through Ekmanpumping in the northern AS and through upwelling in-duced by divergence of the alongshore wind stress com-ponent along the southern India coast during summermonsoon. The detrainment blooms, in initially nutrient-rich but irradiance-constrained water, are attributed to:(1) a gradual build up of nutrients for several months priorto the blooming event, (2) advection of nutrients fromwestern sectors of the AS toward the central AS, and(3) shoaling of the mixed layer. These blooms were iden-tified in the northwestern and southwestern AS. The en-trainment blooms are caused primarily by vertical entrain-ment of nutrients and by secondary sources such as recy-cling after the blooms onset via remineralization of de-tritus.

4.3 Satellite-based observationsAlthough satellite observations relate to only the

upper part of the euphotic zone, near-surface periodicityof phytoplankton, if caused by hydrographic processes,is bound to reflect events involving the entire mixed layer.Based on this rationale, ocean color data from the CZCS,OCM and Sea viewing Wide Field-of-view Sensor(SeaWiFS) have been a central part of the following stud-ies.

Using CSCZ Ch-a data from November 1978 throughDecember 1981, Lierheimer and Banse (2002) examinedseasonal and interannual variability of phytoplankton pig-ment, between 5 and 16°N, and pointed out thatphytoplankton blooms promoted by upwelling duringsummer monsoon are restricted to the shelf, and, on afew occasions, the phytoplankton blooms do occur intothe Lakshadweep Sea and to the south of India. Based onthe analysis of time-sequence CZCS images, they ruledout indigenous phytoplankton blooms in the nutrient-poorLakshadweep Sea. A similar inference was provided byBanse and English (1994) using global maps of Ch-a

medians. They considered medians rather than temporalmeans because the histograms of the monthly means werenon-Gaussian, exhibiting tails with extreme Ch-a values.Their climatological median values indicated a highestpeak (0.9 mg m–3) during September and a secondary peak(0.3 mg m–3) during February at 13°N, 66°E. Emphasiz-ing on Ch-a variability in different sectors of the AS,Banse and English (2000) examined geographical differ-ences in seasonality of CSCZ-derived phytoplankton pig-ment values during 1978–86. For the sector lying to thenorth of 17.5°N and east of 62°E, which forms part ofour review area, Ch-a mean values varied from 0.5 to 1.5mg m–3 with large standard deviation (indicating blooms)on a time scale of 2–8 weeks during late winter monsoon.With a gradual surface warming and with enhanced strati-fication, a decrease in the Ch-a concentration to less than0.2 mg m–3 was inferred during early June.

The seasonal Ch-a variability during 1978–86 for theGulf of Mannar has been described using 1-km spatialresolution Ch-a maps derived from CSCZ (Yapa, 2000).Yapa noted that Ch-a concentrations were higher duringthe summer monsoon than the winter monsoon. Upwellingwas considered as the principle mechanism for high Ch-a production. Recently, composite maps of SST and Ch-afrom OCM have also been employed to provide fisheryforecasts along Gujarat coast based on the mapping ofimportant oceanic features (Solanlki et al., 2001). Initialresults indicated that the forecasts were effective in en-hancing the fish catch 2–3 fold. They opined that a closecoupling between physical and bio-chemical processescould be further explored to locate potential fishing zones.

4.4 Physical mechanisms and primary productionTo gain insights into the seasonal modulation of PP

and Ch-a in relation to mixed layer dynamics and lightpenetration, we present the monthly progression of mixedlayer depth (MLD), photic depth, PP rate integrated forthe photic zone and surface Ch-a estimated from CSCZ(Fig. 9). These parameters were averaged for the coastalregion along the WCI or West India Coastal Province re-ferred to by Longhurst (1998). MLD is the depth wherethe water density is lowered by 0.125σt from the surfaceσt. The photic depth represents the depth where the lightintensity falls by 1% of surface irradiance for water typesI–III (Jerlov, 1964). PP rate (Pt) and Ch-a averaged forphotic zone were estimated from CSCZ data using thealgorithm of Sathyendranath et al. (1995).

MLD shoals due to coastal upwelling in response tospin-up of the southwest monsoon and the pycnocline isilluminated from March to October (Fig. 9, upper panel).Vertical flux of nutrients into the surface mixed layerenhances PP and Ch-a, which show peak values of 70g C m–2·month and 21 mg m–2 during September and Au-gust, respectively (Fig. 9, lower panel). A decrease in the


Ch-a and PP after September is attributed to a buildup ofherbivore population. A weak secondary maximum inMarch in PP and Ch-a is a reflection of wintertime con-vective cooling and deepening in the MLD.

5. Discussion and SummaryThe researches summarized in this review indicate

that the characteristic monsoon-ocean interaction is im-portant in driving and modulating the strength of coastalcurrents, evolution of SST, eddy formation, which even-tually modulates the PP. With the support of the researchwork outlined in the previous sections, we now attemptto seek answers to following issues.

What is the characterization of the annual cycle ofsurface forcing parameters? A temporal march of net heatflux (Qn, negative (positive) represents sea surface cool-ing (warming)), wind speed (W), SST, and wind stresscurl (available for only four months of the year, henceshown by open circles), derived from Hastenrath andLamb (1979a, b), are shown in Fig. 10. Each point repre-sents a 1° × 1° average for box (a), (b) and (c) identifiedin Fig. 1(a). For the northern box, Qn and SST exhibitbimodal oscillations, with peak SST lagging Qn by amonth; W shows only one peak during July (Fig. 10(a)).The curl is highly positive (4 × 10–8 dynes cm–3) duringJuly.

In the vicinity of the central WCI, just south of 15°N,

bimodal oscillations in Qn and SST are also evident, butthe amplitude is almost half of that in the northern box;moreover, the undulations in SST profile are marginal,with an annual mean of ~28°C. Winds are fairly weakduring the first and last fours months of the year; they arestrengthened in July (Fig. 10(b)). Moving southward,while the Qn profile is positive throughout the year, thewind speed displays an interesting pattern: high windspeed (6 m s–1) occurs during June to September, withnodes of weak magnitude (<2 m s–1) during inter-monsoonperiods (Fig. 10(c)). In brief, the annual cycle of surfaceforcing reveals the following. Qn and SST show a bimo-dal cycle along the coast, with decreasing amplitude to-wards the south. Wind and its curl, which is positive,weaken southward with a peak in July. The coastal re-gions are conducive to warming during January to Apriland during post-summer monsoon periods.

AS is a wind-driven basin with seasonally varyingcoastal currents. Coastal upwelling is a consequence ofthe alongshore wind stress component all year round alongthis coast. So, what is the seasonal upwelling scenarioalong the coast? With the monthly mean alongshore windstress component directed toward the equator (Shetye etal., 1985), divergence of the alongshore wind-stress would

Fig. 10. Climatological seasonal march of wind speed (W),surface net heat flux (Qn), sea surface temperature (SST),and wind stress curl which is represented by open circles asit was available for only January, April, July and October.Area-mean was taken on each of the 1° × 1° boxes shownin Fig. 1(a). The centered locations are: (a) 23.5°N, 66.5°E,(b) 14.5°N, 73.5°E and (c) 6.5°N, 77.5°E. Data source:Hastenrath and Lamb (1979a, b).

Fig. 9. Climatological seasonal march of mixed layer depth(MLD), photic depth, primary production rate integrated forphotic zone (Pt) and surface chlorophyll-a (Ch-a) fromCoastal Zone Color Scanner. Source: Longhurst (1998).


promote upwelling throughout the year. In fact,hydrographic data suggest that the upwelling starts inMarch and continues till September along the Kerala coast(Johannessen et al., 1981).

What is the relation between monsoon forcing andPP? PP along the southern India coast is enhanced duringthe summer monsoon due to an increase in the upwellingintensity promoted by offshore flux of surface water inresponse to equatorward alongshore wind stress compo-nent. The northern AS supports high PP in response toupward Ekman pumping promoted by positive wind stresscurl induced by the Findlater jet (Fig. 1(a)). The scenarioduring the winter monsoon is dominated by upper-oceanconvective cooling and deepening of MLD to ~100 m(Rao et al., 1989) and vertical influx of nutrients into theeuphotic layer, which together with an increase irradi-ance enhance PP on an annual scale (Qasim, 1982;Madhupratap et al., 1996). In the context of local forc-ing, the monsoons modulate the geographical distribu-tion of PP and Ch-a in the eastern AS.

In summary, in this review we synthesizemultidisciplinary aspects, ranging from air-sea interac-tion to PP, based on a number of investigations publishedin international and Indian journals. The intention is toprovide a condensed, illustrative summary of these in-vestigations that would serve as a guide for future works.It should also provide insights on what has been exploredand what still needs to be probed. Currently, there areeither no answers or only partial answers to the follow-ing questions. (1) What is the coupling between AS SSTand rainfall over the Indian subcontinent? (2) What is theseasonal pattern of salinity along the eastern AS? (3) Whatis the seasonal cycle of Ch-a distribution in EEZ of theIndian coast? (4) What is role of El Niño-Southern Oscil-lation on coastal SST? (5) What is the role of shelf to-pography on the three-dimensional circulation? A majorobstacle in addressing these issues is the lack of data ofsufficient spatial and temporal resolution. With more thana dozen satellites in space and with the operational satel-lite oceanography in place, and with enhanced comput-ing systems, it would be tempting to attempt these andmany unresolved issues relating to the dynamics of thishighly productive and important sector of the north In-dian Ocean.

AcknowledgementsThis study was supported by MEXT RR2002 Project

for Sustainable Coexistence of Human, Nature and Earth(category 7). Additional support from ADEOS-I and IIprojects of NASDA, Japan, is also acknowledged.

AcronymsAS Arabian SeaASHSW Arabian Sea High Salinity Water

ASW Arabian Sea WaterAVHRR Advanced Very High Resolution RadiometerBBW Bay of Bengal WaterCh-a Chlorophyll-aCTD Conductivity Temperature DepthCZCS Coastal Zone Color ScannerEEZ Exclusive Economic ZoneEICC East India Coastal CurrentESW Equatorial Surface WaterE-P Evaporation minus precipitationGM Gulf of MannarIEW Indian Equatorial WaterIIOE International Indian Ocean ExpeditionIT Indian tipJGOFS Joint Global Ocean Flux StudiesLH/LL Lakshadweep High/LowMLD Mixed Layer DepthNASA National Aeronautics and Space AdministrationNCEP/NCAR National Centers for Environmental Prediction/

National Center for Atmospheric ResearchNMC Northeast Monsoon CurrentNSCAT NASA ScatterometerOCM Ocean Color MonitorPGW Persian Gulf WaterPOM Princeton Ocean ModelPP Primary ProductionPt Primary Production rateQn Surface net heat fluxRHJ Ras al Hadd JetRSPGW Red Sea Persian Gulf Intermediate WaterRSW Red Sea WaterSD Sri Lanka DomeSeaWiFS Sea viewing Wide Field of view SensorSMC Southwest Monsoon CurrentSSM/I Special Sensor Microwave ImagerSST Sea Surface TemperatureTCZ Tropical Convergence ZoneUNDP/FAO United Nations Development Program/Fishery

Assessment OfficeWCI West Coast of IndiaWICC West India Coastal Current

ReferencesBanse, K. (1968): Hydrography of the Arabian Sea shelf of In-

dia and Pakistan and effects on dimersal fishes. Deep-SeaRes., 15, 45–79.

Banse, K. (1987): Seasonality of phytoplankton chlorophyll inthe central and northern Arabian Sea. Deep-Sea Res., 34,713–723.

Banse, K. and D. C. English (1994): Seasonality of coastal zonecolor scanner phytoplankton pigment in the offshore oceans.J. Geophys. Res., 99, 7323–7346.

Banse, K. and D. C. English (2000): Geographical differencesin seasonality of CSCZ-derived phytoplankton pigment inthe Arabian Sea for 1978–1986. Deep-Sea Res., 47, 1623–1677.

Bauer, S., G. L. Hitchcock and D. B. Olson (1991): Influenceof monsoonally-forced Ekman dynamics upon surface layerdepth and plankton biomass distribution in the Arabian Sea.


Deep-Sea Res., 38(5), 531–553.Brock, J., S. Sathyendranath and T. Plat (1994): A model study

of seasonal mixed-layer primary production in the ArabianSea. p. 65–78. In Biochemistry of the Arabian Sea, ed. byD. Lal, Ind. Acad. Sci.

Bruce, J. G., D. R. Johnson and J. C. Kindle (1994): Evidencefor eddy formation in the eastern Arabian Sea during thenortheast monsoon. J. Geophys. Res., 99, 7651–7664.

Bruce, J. G., J. C. Kindle, L. H. Kantha, J. L. Kerling and J. F.Bailey (1998): Recent observations and modeling in theArabian Sea Laccadive High region. J. Geophys. Res., 103,7593–7600.

Central Marine Fisheries Research Institute (1995): Mar. Fish.Infor. Serv., Technical and Extension Series, Vol. 136, 30pp.

Chauhan, P., C. R. C. Nagur, M. Mohan, S. R. Nayak and R. R.Navalgund (2001): Surface chlorophyll-a distribution inArabian Sea and Bay of Bengal using IRS-P4 Ocean ColorMonitor satellite data. Curr. Sci., 80, 127–129.

Colborn, J. (1976): The Thermal Structure of the Indian Ocean.The University Press of Hawaii, 173 pp.

Cutler, A. N. and J. C. Swallow (1984): Surface currents of theIndian Ocean (to 25°S, 100°E). Complied from historicaldata archived by the Meteorological Office, Bracknel, UK,Rep. 187, Institute of Oceanographic Sciences, Wormley,UK. 8 pp. and 36 charts.

Dabyshire, M. (1967): The surface waters off the coast ofKerala, southwest India. Deep-Sea Res., 14, 295–320.

Düing, W. and A. Leetmaa (1980): Arabian Sea cooling: A pre-liminary heat budget. J. Phys. Oceanogr., 10, 307–312.

Emery, W. J. and J. Meincke (1986): Global water masses: sum-mary and review. Oceanologica Acta, 9, 383–391.

Findlater, J. (1977): Observational aspects of the low-levelcross-equatorial jet stream on the western Indian Ocean.Pure Appl. Geophys., 115, 1251–1262.

Gadgil, S. (2000): Monsoon-ocean coupling. Curr. Sci., 78, 309–322.

Gupta, R. S., A. Branganza, R. J. Noronha and S. Y. S. Singbal(1980): Chemical oceanography of the Arabian Sea: PartIV, Hydrochemical characteristics off the central west coastof India. Indian J. Mar. Sci., 9, 240–245.

Hastenrath, S. and P. J. Lamb (1979a): Climatic Atlas of theIndian Ocean, Part 1, Surface Climate and AtmosphericCirculation. Wisconsin University Press, Madison, 97charts, 19 pp.

Hastenrath, S. and P. J. Lamb (1979b): Climatic Atlas of theIndian Ocean, Part 2, The Oceanic Heat Budget. Wiscon-sin University Press, Madison, 83 charts, 15 pp.

Haugen, V. E., O. M. Johannessen and G. Evensen (2002):Mesoscale modeling study of the oceanographic conditionsoff the southwest coast of India. Proceed. Ind. Acad. Sci.(Earth and Planet. Sci.), 111, 321–337.

Jerlov, N. G. (1964): Optical classification of ocean water.p. 45–49. In Physical Aspects of Light in the Sea, Univ. Ha-waii Press, Honolulu.

Johannessen, O. M., G. Subbaraju and J. Blindheim (1981):Seasonal variation of the oceanographic conditions off thesouthwestern coast of India during 1971–75. FiskDir. Skr.Ser. HavUnders., 18, 247–261.

Kalnay, E., M. Kanamitsu, R. Kistler, W. Collins, D. Deaven,L. Gandin, M. Iredell, S. Saha, G. White, J. Woollen, Y.Zhu, M. Chelliah, W. Ebisuzaki, W. Higgins, J. Janowiak,K. C. Mo, C. Ropelwski, J. Wang, A. Leetmaa, R. Reynolds,R. Jenne and D. Joseph (1996): The NCEP/NCAR 40-yearreanalysis project. Bull. Ame. Meteorol. Soc., 77, 437–471.

Krishnamurti, T. N., P. Ardanuy, Y. Ramanathan and R. Pasch(1981): On the onset of vortex of the summer monsoon.Mon. Wea. Rev., 109, 344–363.

Kumar, S. P. and T. G. Prasad (1996): Winter cooling in thenorthern Arabian Sea. Curr. Sci., 71, 834–841.

Kumar, S. P. and T. G. Prasad (1999): Formation and spreadingof Arabian Sea high salinity water mass. J. Geophys. Res.,104, 1455–1464.

Kumar, S. P., M. Madhupratap, M. D. Kumar, M. Gauns, P. M.Muraleedharan, V. V. S. S. Sarma and S. N. DeSouza (2000):Physical control of primary productivity on a seasonal scalein central and eastern Arabian Sea. Proceed. Ind. Acad. Sci.(Earth and Planet. Sci.), 109, 433–441.

Lierheimer, L. J. and K. Banse (2002): Seasonal and interannualvariability of phytoplankton pigment in the Laccadive(Lakshadweep) Sea as observed by the Coastal Zone ColorScanner. Proceed. Ind. Acad. Sci. (Earth and Planet. Sci.),111, 163–185.

Longhurst, A. R. (1998): Ecological Geography of the Sea.Academic Press, San Diego, California, U.S.A., 398 pp.

Luis, A. J. and H. Kawamura (2000): Wintertime wind forcingand sea surface temperature cooling near the south Indiatip observed using NSCAT and AVHRR. Remote Sens.Environ., 73, 55–64.

Luis, A. J. and H. Kawamura (2001): Characteristics of atmos-pheric forcing and SST cooling events in the Gulf of Mannarduring winter monsoon. Remote Sens. Environ., 77, 1–10.

Luis, A. J. and H. Kawamura (2002a): Dynamics and mecha-nism for sea surface cooling near the Indian tip during win-ter monsoon. J. Geophys. Res., 107, 3187, doi: 10.1029/2000JC000455.

Luis, A. J. and H. Kawamura (2002b): A case study of SSTcooling dynamics near the Indian tip during May 1997. J.Geophys. Res., 3171, doi: 10.1029/2000JC000778.

Madhupratap, M., S. P. Kumar, P. M. A. Bhattathiri, M. D.Kumar, S. Raghukumar, K. K. C. Nair and N. Ramaiah(1996): Mechanism of the biological response to wintercooling in the northeastern Arabian Sea. Nature, 384, 549–552.

McCreary, J. P. and P. K. Kundu (1989): A numerical investi-gation of sea surface temperature variability in the ArabianSea. J. Geophys. Res., 94, 16097–16114.

McCreary, J. P., S. R. Shetye and P. K. Kundu (1986):Thermohaline forcing of eastern boundary currents: withapplication to the circulation off the west coast of Australia.J. Mar. Res., 44, 71–92.

McCreary, J. P., P. K. Kundu and R. L. Molinari (1993): A nu-merical investigation of dynamics, thermodynamics andmixed-layer processes in the Indian Ocean. Prog. Oceanogr.,31, 181–224.

McCreary, J. P., K. E. Kohler, R. R. Hood and D. B. Olson(1996): A four-component ecosystem model of biologicalactivity in the Arabian Sea. Prog. Oceanogr., 37, 193–240.


Muraleedharan, P. M. and S. P. Kumar (1996): Arabian Seaupwelling, a comparison between coastal and open oceanregions. Curr. Sci., 71, 842–846.

Naidu, P. D., M. R. R. Kumar and V. R. Babu (1999): Time andspace variations of monsoonal upwelling along the west andeast coasts of India. Cont. Shelf Res., 19, 559–572.

Prasad, T. G. (1996): Mixed layer depth and thermocline cli-matology of the Arabian Sea and western equatorial IndianOcean. Indian J. Mar. Sci., 25, 189–194.

Prasad, T. G. (1997): Annual and seasonal mean buoyancy fluxesfor the tropical Indian Ocean. Curr. Sci., 73, 667–674.

Qasim, S. Z. (1982): Oceanography of the northern ArabianSea. Deep-Sea Res., 29, 1041–1068.

Rao, R. R., R. L. Molinari and J. F. Fiesta (1989): Evolution ofthe climatological near-surface thermal structure of thetropical Indian Ocean, Part 1: Description of mean monthlymixed-layer depth and sea-surface temperature, surfacecurrent and surface meteorological fields. J. Geophys. Res.,94, 10801–10815.

Rochford, D. J. (1964): Salinity maxima in the upper 1000 mof the North Indian Ocean. Australian J. Mar. FreshwaterRes., 15, 1–24.

Sastry, J. S. and R. S. D’Souza (1972): Oceanography of theArabian Sea during southwest monsoon season. Part III:Salinity. Indian J. of Meteorol. and Geophys., 23, 479–490.

Sathyendranath, S., A. R. Longhurst, C. M. Caverhill and T.Platt (1995): Regionally and seasonally differentiated pri-mary production in the north Atlantic. Deep-Sea Res., 42,1773–1802.

Sawant, S. and M. Madhupratap (1996): Seasonality and com-position of phytoplankton in the Arabian Sea. Curr. Sci.,71, 869–873.

Schott, F. (1983): Monsoon response of the Somali Current andassociated upwelling. Prog. Oceanogr., 12, 357–381.

Shankar, D. (2000): Seasonal cycle of sea level and currentsalong the coast of India. Curr. Sci., 78, 279–288.

Shankar, D. and S. R. Shetye (1997): On the dynamics of theLakshadweep high and low in the southeastern Arabian Sea.J. Geophys. Res., 102, 12551–12562.

Shankar, D. and S. R. Shetye (1999): Are interdecadal sea levelchanges along the Indian coast influenced by variability ofmonsoon rainfall? J. Geophys. Res., 104, 26031–26042.

Shankar, D., P. N. Vinayachandran and A. S. Unikrishnan(2002): The monsoon currents in the north Indian Ocean.Prog. Oceanogr., 52, 63–120.

Shenoi, S. S. C., D. Shankar and S. R. Shetye (1999): On thesea surface temperature high in the Lakshadweep Sea be-fore the onset of the southwest monsoon. J. Geophys. Res.,104, 15703–15712.

Shetye, S. R. (1984): Seasonal variability of the temperaturefield off the southwest coast of India. Proceed. Ind. Acad.Sci. (Earth and Planet. Sci.), 93, 399–411.

Shetye, S. R. (1986): A model study of the seasonal cycle ofthe Arabian Sea surface temperature. J. Mar. Res., 44, 521–542.

Shetye, S. R. and S. C. C. Shenoi (1988): Seasonal cycle ofsurface circulation in the coastal north Indian Ocean. Pro-ceed. Ind. Acad. Sci. (Earth and Planet. Sci.), 97, 53–62.

Shetye, S. R., S. S. C. Shenoi, M. K. Antony and V. K. Kumar(1985): Monthly-mean wind stress along the coast of thenorth Indian Ocean. Proceed. Ind. Acad. Sci. (Earth andPlanet. Sci.), 94, 129–137.

Shetye, S. R., A. D. Gouveia, S. C. C. Shenoi, D. Sundar, G. S.Michael, A. M. Almeida and K. Santanam (1990):Hydrography and circulation off the west coast of Indiaduring the southwest monsoon 1987. J. Mar. Res., 48, 359–378.

Shetye, S. R., A. D. Gouveia, S. S. C. Shenoi, G. S. Michael,D. Sundar, A. M. Almeida and K. Santanam (1991): Thecoastal current off western India during the northeastmonsoon. Deep-Sea Res., 38, 1517–1529.

Solanlki, H. U., R. M. Dwivedi, S. R. Nayak, J. V. Jadeja, D. B.Thakar, H. B. Dave and M. I. Patel (2001): Application ofOcean Color Monitor chlorophyll and AVHRR SST for fish-ery forecast, preliminary validation results off Gujarat coast,northwest coast of India. Indian J. Mar. Sci., 30, 132–138.

Stramma, L., J. Fisher and F. Schott (1996): The flow field offthe southwest India at 8°N during the southwest monsoonof August 1993. J. Mar. Res., 54, 55–72.

Sverdrup, H. U., M. W. Johnson and R. H. Fleming (1942): TheOceans: Their Physics, Chemistry and Biology. Prentice-Hall, New York, 1087 pp.

Swallow, J. C. (1984): Some aspects of the physical oceanog-raphy of the Indian Ocean. Deep-Sea Res., 31, 639–650.

Tomczak, M. and J. S. Godfrey (1994): Regional Oceanogra-phy, An Introduction. Pergamon Press, Butler and TannerLtd., Great Britain, 422 pp.

World Ocean Atlas-1998 (1999): CDROM documentation, ver.1, Ocean Climate Laboratory, National Oceanographic DataCenter, Silver Spring, MD, 16 pp.

Wyrtki, K. (1971): Oceanographic Atlas of the InternationalIndian Ocean Expedition. U.S. Govt. Printing Office, Wash-ington, D.C., 531 pp.

Wyrtki, K. (1973): Physical oceanography of the Indian Ocean.p. 18–36. In The Biology of the Indian Ocean, EcologicalStudies, 3, ed. by B. Zeitzschel and S. A. Gerlach, Springer-Verlag, London.

Yapa, K. K. A. S. (2000): Seasonal variability of sea surfacechlorophyll-a of waters around Sri Lanka. Proceed. Ind.Acad. Sci. (Earth and Planet. Sci.), 109, 427–432.

Yu, L., J. J. O’Brien and J. Yang (1991): On the remote forcingof the circulation in the Bay of Bengal. J. Geophys. Res.,96, 20449–20452.

Documents

Air-Sea Interaction, Coastal Circulation and Primary ...svr4.terrapub.co.jp/journals/JO/pdf/6002/60020205.pdfduces upwelling which lowers sea surface temperature by about 2.5°C (compared