(This is a sample cover image for this issue. The actual cover is … · 2019. 6. 24. · in Fig....

(This is a sample cover image for this issue. The actual cover is not yet available at this time.)

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright

http://www.elsevier.com/copyright

Author's personal copy

Stable isotopes and salinity in the surface waters of the Bay of Bengal: Implications forwater dynamics and palaeoclimate

Hema Achyuthan a, R.D. Deshpande b,⁎, M.S. Rao c, Bhishm Kumar c,1, T. Nallathambi d, K. Shashi Kumar d,R. Ramesh e, P. Ramachandran e, A.S. Maurya b,2, S.K. Gupta b

a Department of Geology, Anna University, Chennai 600 025, Indiab Physical Research Laboratory, Navrangpura, Ahmedabad 380 009, Indiac National Institute of Hydrology, Roorkee, Uttarakhand 247 667, Indiad National Institute of Ocean Technology, Pallikaranai, Chennai 600 025, Indiae Institute of Ocean Management, Anna University, Chennai 600 025, India

a b s t r a c ta r t i c l e i n f o

Article history:Received 25 May 2012Received in revised form 27 December 2012Accepted 27 December 2012Available online 5 January 2013

Keywords:Bay of BengalStable isotopesSalinitySurface watersHimalayan riversPeninsular riversIndia

Surface water mixing in the Bay of Bengal (BOB) inferred from spatio-temporal distribution of δ18O andsalinity based on synthesis of 194 new samples together with published data is reported. In general, bothδ18O and salinity have low values in northern part of the BOB, progressively increasing towards SW. Thelowest values are observed during July–September (southwest monsoon season) and the highest in pre-monsoon. The most prominent δ18O–salinity relationship is seen for samples collected during June to Octoberwhen the Himalayan river influx dominates. When this influx decreases in other seasons the δ18O–salinityrelationship is poor.The δ18O–δD regression of samples north of 10°N is similar to the GMWL. However, for samples south of~10°N, this regression has a significantly lower slope. This is interpreted as due to absence of direct riverineinflow in this region of the BOB, coupled with –ve (P−E; Precipitation minus Evaporation) almost through-out the year.This study shows that the seasonal distribution of δ18O and salinity over the northern BOB is dominantlygoverned by the variation in the (P+R− |E|) in spite of the fact that ocean currents transfer several timesmore water between the two basins of northern Indian Ocean.

© 2013 Elsevier B.V. All rights reserved.

1. Introduction

Studies on the dynamics of surface water of the northern IndianOcean are important as they significantly affect monsoon systematicsand consequently the water availability over India. A complex inter-play of several factors/processes which include river influx andocean currents, direct precipitation and evaporation govern thesedynamics. Natural or anthropogenic perturbations in any of thesecontrolling factors/processes can have far reaching and non-linearconsequences to meteorology, atmospheric chemistry and heat bud-get of the region. Study of tracers such as oxygen and hydrogen isoto-pic ratios (δ18O and δD) that essentially track the water molecules,and salinity are best suited for studying the involved dynamicalprocesses.

The BOB is a unique ocean basin influenced by seasonally reversingmonsoon winds, high precipitation [~2 m/year; (Prasad, 1997)] andlarge influx of freshwater from Himalayan and Peninsular Rivers ofthe Indian sub-continent. The geographical location of the BOB andthe regions where major rivers discharge freshwater into it are shownin Fig. 1a. The monthly distribution of precipitation (P), runoff (R),evaporation (E) and sea surface salinity (SSS), is shown in Fig. 1b (Raoand Sivakumar, 2003). While the maximum of precipitation occurs inmonths of June–July, most of the runoff occurs in June through Novem-ber. The evaporation on the other hand is bimodal with a maximum inMarch–April and another high in September–October. The maximumvalue of (P+R− |E|) occurs in the month of July–August–Septemberconsequently the minimum values of SSS are observed in August andSeptember. The SSS progressively increases through November until asecond minimum occurs in December in response to winter monsoonover Peninsular India.

The total freshwater discharge from the major rivers is estimated tobe ~1630 km3a−1 based on data from gauge stations (Biksham andSubramanian, 1980; Chakrapani and Subramanian, 1990; Martin et al.,1981; Ramesh and Subramanian, 1993; Subramanian, 1993). Monthlydistributions of river discharges of the major rivers draining into the

Marine Chemistry 149 (2013) 51–62

⁎ Corresponding author. Tel.: +91 7926314256; fax: +91 7926314900.E-mail address: [email protected] (R.D. Deshpande).

1 Presently at International Atomic Energy Agency, Vienna.2 Presently at Department of Earth Sciences, Indian Institute of Technology, Roorkee

247 667, India.

0304-4203/$ – see front matter © 2013 Elsevier B.V. All rights reserved.http://dx.doi.org/10.1016/j.marchem.2012.12.006

Contents lists available at SciVerse ScienceDirect

Marine Chemistry

j ourna l homepage: www.e lsev ie r .com/ locate /marchem


BOB are shown in Fig. 1c. Maximum river discharge occurs duringJune to November peaking in August. The southern Peninsular Rivers(Cauvery, Mahanadi, Godavari and Krishna) contribute limited riverdischarge to the BOB in comparison to the major rivers draining theHimalayan region (Brahmaputra, Ganges and Irrawaddy).

In addition to major rivers that are gauged, numerous smallerstreams also discharge into the BOB. The total annual continental run-off along the BOB coast line north of 6°N was estimated at 2950 km3

which is about 60% of the total runoff into the entire tropical IndianOcean north of 30°S (Sengupta et al., 2006). The annual total precip-itation and evaporation over the BOB are estimated to be 4700 km3

and 3600 km3 respectively (Sengupta et al., 2006), resulting in over-all positive (P−E) for the BOB. However, during summer, (P−E) isnegative as discussed subsequently. But, due to large river discharges,(P+R− |E|) is positive even during summer. As a result, the upperlayers of the BOB have lower salinity (by 3–7), and warmer seasurface temperature (by 1.5−2 °C) than the ocean basin to the westof the Indian land mass, namely, the Arabian Sea (AS) (PrasannaKumar et al., 2002). During this period, winds are unable to breakthe strongly stratified surface layer of the BOB; thereby restrictingthe turbulent wind-driven vertical mixing to a shallow depth ofb20 m (Prasanna Kumar et al., 2002). This stratified layer also re-stricts exchange of heat between the deeper layer and the atmo-sphere, maintaining sea surface temperature (SST) in BOB >28 °C.This phenomenon supports large scale deep convection in the atmo-sphere during the summer monsoon (Shenoi et al., 2002). It hasbeen calculated that SST of 28–29 °C is needed for charging thecloud-base air-mass with the required moist static energy for cloudsto reach the upper troposphere, i.e. ~200 hPa (Gadgil, 2003; Gadgilet al., 1984; Sud et al., 1999).

As with the winds (Fig. 2a), the oceanic currents (Fig. 2b) in thenorthern Indian ocean also reverse from summer to winter. The majorocean surface currents that flow between the AS and the BOB are: (i)The Summer Monsoon Current (SMC), flowing eastward during Mayto September; and (ii) the Winter Monsoon Current (WMC), flowingwestward during November to February. These ocean currents extendover the entire northern Indian Ocean from the Somali coast to the east-ern BOB but they arise or decay over thiswhole region at different times.Only in their mature phase these currents exist as trans-basin flows(Shankar et al., 2002). The eastward flowing SMC first appears in thesouthern BOB during May. In its mature phase, peaking with the sum-mer monsoon in July, the SMC in the AS is a continuation of the SomaliCurrent and the coastal current of Oman. The SMC flows eastward southof Sri Lanka and into the BOB (Fig. 2b; upper panel) as East India CoastalCurrent (EICC). The westward WMC first forms south of Sri Lanka inNovember and is fed mainly by the equatorward East India Coastal Cur-rent (EICC). In its mature phase during December to March, the WMCflows westward across the southern BOB (Fig. 2b; lower panel) and di-vides into two branches in the AS. One of these branches continuesflowing westward whereas the other turns around to flow northwardsalong the western coast of India and is known as West India CoastalCurrent (WICC). It is thus seen that EICC is a western boundary currentin the BOB, along east coast of India that reverses with monsoon.Shankar et al. (2002) indicated that SMC and WMC transport~10×106m3s−1 (i.e. ~10 Sverdrup or ~3×105km3a−1) of water eitherway, in the upper 400 m, with most of the transport being restricted toupper 100 m. This transport of water is more than 100 times the totalriver discharge into the BOB.

Another important source of water into the southern Indian Oceanis the Indonesian Throughflow (ITF) that provides large amount (10

Fig. 1. (a) The geographical locations of the regions in the BOB where major rivers from Peninsular India (Cauvery, Godavari, Krishna and Mahanadi) and Himalaya (Ganga,Brahmaputra and Irrawaddy) discharge freshwater. The acronyms indicate: IND: India; Ch: Chennai; Vi: Visakhapatnam; Ko: Kolkata; BGD: Bangladesh; MM: Myanmar and Po:Port Blair. (b) Annual cycle of evaporation ‘E’, precipitation ‘P’, river discharges ‘R’ from all the major rivers flowing into the BOB, also shown are ‘P+R− |E|’ and sea surface salinity‘SSS’ for the entire BOB (north of 8°N); redrawn from (Rao and Sivakumar, 2003). (c) Average annual cycle of discharge frommajor rivers in the BOB (Data source: http://www.sage.wisc.edu/riverdata/).

52 H. Achyuthan et al. / Marine Chemistry 149 (2013) 51–62


to 14 sV) of warm and fresh water from the Pacific through Indone-sian Seas (Sprintall et al., 2002; Sprintall et al., 2009; Wijffels et al.,2002). This inter-basin exchange of waters is known to have an im-pact on mass, heat and freshwater budgets of the south Pacific andsouthern Indian Oceans. However, this water mass is introducedinto the BOB in an area south of the 10°S, which is not covered bycruises in this study.

Another important aspect concerning the BOB is the large differ-ence in seasonal exchange of water with the atmosphere in terms of(P−E). During winter monsoon, the oceanic precipitation exceedsevaporation, with limited river discharge (Fig. 2c, lower panel). Incontrast, almost entire northeastern BOB and large part of the ASalong the west coast of India have a negative value of P−E duringsummer monsoon (Fig. 2c; upper panel), indicating that these partscontribute significant amount of vapor to the summer monsoonrains in India. However, as mentioned earlier, due to discharge fromHimalayan Rivers, the (P+R− |E|) has a large positive value andleads to lowering of SSS in the BOB during this period (Fig. 1b).

Most studies on the BOB surface water dynamics so far involvedcruise based analyses of physical and chemical properties or satellitebased observations and modeling. In addition, some investigatorshave also studied δ18O in the BOB surface and deep waters and its re-lationship with salinity (S) (Delaygue et al., 2001; Singh et al., 2010;Somayajulu et al., 2002; Srivastava et al., 2010, 2007). Different oceansmay have characteristic δ18O–S relationships depending on physicalprocesses, namely, precipitation, evaporation, river influx, upwelling,advection, dispersion and mixing of different water masses involved(Bigg and Rohling, 2000; Ferronsky and Brezgunov, 1989; Singh et al.,2010). Since these processes vary both spatially and temporally, theδ18O–S relationships often provide useful information about operatingphysical processes (Benway and Mix, 2004).

The δ18O–S relationship is also used in palaeoclimatology to esti-mate past sea surface salinity variations and for understanding thepast ocean dynamics. The past-to-present δ18O variations are inferredfrom study of foraminifera shells in ocean sediments. However, δ18Ovariations of foraminiferal tests include contributions from sea surfacetemperatures, global ice volume and local salinity, which may be dif-ficult to separate. Recently a new technique for deriving palaeo-temperatures has been developed that is based on abundance ratio ofunsaturated alkenone in phytoplankton algae in sediments (Brassellet al., 1986; Prahl andWakeham, 1987). Thus, combined use of oxygen

isotopes and alkenone records can give information on temperatureand the corresponding compensation to δ18O record for estimating sa-linity in the past (Rostek et al., 1993).

The salinity reconstruction is based on the slope of the δ18O–S rela-tionship, assuming that past-to-present variation (Δδ18O/ΔS) can beinferred from the observed present day spatial slope of the δ18O–S rela-tionship (Delaygue et al., 2001). Improved understanding about thespatio-temporal variation in the present time δ18O–S relationship inthe BOB is thus valuable for reliable palaeoclimatic inferences about sa-linity variation inferred from δ18O of foraminifera, and for appreciatingthe limitation of such reconstruction.

Although the application of δ18O and δ18O–S relationship as a tool tounderstandmarine processes is well known, the available data from theBOB, is limited for delineating these processes (Abe et al., 2009;Delaygue et al., 2000; Strain and Tan, 1993). Prior to this study, thereexisted only 196 data pairs (δ18O and S) pertaining to surface watersof the BOB (Delaygue et al., 2001; Singh et al., 2010; Somayajulu et al.,2002). Compared to the large surface area (~2.2×106km2) of theBOB, the available data are too few to derive reliable trends of seasonaland inter-annual variations in δ18O and S. Further, some of the currentinformation on δ18O–S relationship in surface water of the BOB isbased on indirect estimates of δ18O derived from foraminifera shells, as-suming their precipitation under isotopic equilibrium (Duplessy, 1982;Duplessy et al., 1981; Singh et al., 2010), and S derived from δ18O andalkenone records (Rostek et al., 1993) of sediment cores.

The alkenone based reconstruction of past sea surface temperaturehas been reported to agree well with sea surface temperature basedon foraminifera assemblages from the tropical Indian Ocean (Clemenset al., 1991; Prell and Hutson, 1979; Rostek et al., 1993). However, itmust be noted that the study combining δ18O and alkenone records indeep-sea core to extract the salinity signal was undertaken in the IndianOcean at the juncture (~5°N) of the AS and the BOB, however, reliableSST and past salinity reconstructions in the two basins are still not avail-able. In view of such limited data availability at present, the need formore extensive paired isotope and salinity dataset from the BOB wasstrongly felt to investigate their spatial distribution and linkages interms of controlling factors.

To partially achieve this goal, samples from five cruises conductedduring different seasons of 2007–2008, in the BOB were analyzed, aspart of a National Programme on Isotope Fingerprinting of Waters ofIndia (IWIN) (Deshpande and Gupta, 2008, 2009, 2012). These cruises

Fig. 2. (a) Climatological wind stress over northern Indian Ocean for the months of July and January. (Redrawn from: (Shankar et al., 2002) (b) Schematic representation of thecirculation across the AS and the BOB during summer and winter (Redrawn from (Shankar et al., 2002)). (c) Precipitation minus Evaporation during summer and winter monthsin the AS and the BOB. (Redrawn from (Yu and Weller, 2007).

53H. Achyuthan et al. / Marine Chemistry 149 (2013) 51–62


were undertaken using either ships of the National Institute of OceanTechnology (NIOT) or commercial shipping vessels.

2. Methodology

The sea water samples in this study were collected during thefollowing five cruises: (i) Chennai to Andaman (December, 2007; com-mercial vessel); (ii) Andaman to Visakhapatnam (March, 2008; com-mercial vessel); (iii) Chennai to Andaman and back (June, 2008; ORVSagarnidhi-1); (iv) Chennai to south of equator (August, 2008; ORVSagarnidhi-2); and (v) Andaman to Kolkata and back (November–December, 2008; commercial vessel). The samples were collectedfrom the bow side of the vessel between the depths of 1 to 10 m andwere transferred into clean and dry containers from the Niskin sampler.The samples were analyzed onboard, for various physico-chemical pa-rameters such as water temperature, dissolved oxygen, salinity, pH,conductivity, turbidity and specific gravity, using a multi-parameterprobe (TOA-DKK Sensor module-WMS-24) with a salinity accuracy of±0.1%. This paper presents only the isotopic measurements of oxygenand hydrogen, and salinity of water samples. The salinity here refersto Practical Salinity (S) defined in terms of the ratio of the electrical con-ductivity of a seawater sample at the temperature of 15 °C and the pres-sure of one standard atmosphere, to that of a KCl solution, in which themass fraction is 32.4356×10−3, at the same temperature and pressure.This ratio equal to 1 corresponds, by definition, to a Practical Salinity ex-actly equal to 35 (Millero, 2011; Millero et al., 2008; UNESCO, 1981a,b).The interpretations in this paper are based on synthesis of publisheddata and those obtained in this study enabling coverage of widespatio-temporal range.

The isotopic analyses of water samples (δ18O and δD) were carriedout at two laboratories, the National Institute of Hydrology (NIH),Roorkee and the Physical Research Laboratory (PRL), Ahmedabad fol-lowing standard equilibration method in which water samples wereequilibrated with CO2 (or H2) and the equilibrated CO2 (or H2) gaswas analyzed in Isotope Ratio Mass Spectrometer (IRMS). At NIH,the δ18O and δD analyses were made using Isoprime IRMS in DualInlet mode (Kumar et al., 2010). At PRL, these analyses were madeusing Delta V Plus IRMS in continuous flow mode using Gasbench II(Maurya et al., 2009). The reproducibility of measurements at boththe laboratories was better than 0.1‰ for δ18O and 1‰ for δD.Inter-laboratory calibration including Isotope Hydrology Laboratoryat IAEA, Vienna indicated that analyses from all the three laboratorieswere in agreement within ±0.1‰ for δ18O and 1‰ for δD.

3. Results and discussions

The seasonal variations (March to May, June to October, andNovember to February) in the geographical distribution of δ18O andsalinity of surface waters of the BOB, along with sampling locations,are shown in Figs. 3, 4 and 5 respectively. The contour plots arebased on synthesis of data from this study (194 samples) and thoseavailable in published literature (Delaygue et al., 2001; Singh et al.,2010). The cruise names and codes as used for plotting the geograph-ical distributions (Figs. 3, 4 and 5) of salinity and δ18O are listed inTable 1. The sampling details along with measured / computed valuesof salinity, δ18O, δD and d-excess are given in Table 2.

The March–May is a transitional period from winter to summermonsoon and oceanic currents are feeble. During this pre-monsoonperiod, water discharges from both the Peninsular and the HimalayanRivers are low. At the same time the air temperatures and SST pro-gressively rise (Shenoi et al., 2002) and evaporation from the BOB ishigh (Fig. 1b) leading to enrichment of both salinity and δ18O in theentire BOB. This is consistent with the highest values of both salinityand δ18O observed during this season (Fig. 3) compared to other sea-sons. Additionally, a general increase of both these parameters fromthe central BOB to southwestern BOB is observed. This is the resultof continuing inflow from the north and northeast from the Himala-yan Rivers, though at a reduced rate. Embedded within this generalincrease of the two parameters, is a large patch of high salinitywater and a relatively smaller patch of high δ18O in the north. Thesepatches, indicating significant evaporation, are enclosed by relativelylower values of both these parameters. The smaller size of the highδ18O patch compared to high salinity patch could be the result ofhigher contrast of δ18O, (vis-à-vis salinity), between sea water andfresh Himalayan influx which is largely snow melt with much lowerδ18O during this season. A low δ18O patch off Chennai is possibly aremnant from the previous season (November–February) seen inFig. 5b and discussed subsequently.

During the summer monsoon season, June to October (Fig. 4), themost prominent change from March–May period is the much lowervalues of δ18O and salinity in the surface waters of the BOB north of~10°N, particularly along the east coast of India. Due to absence ofsampling locations in the NE BOB, the extent of decrease in bothδ18O and salinity in this region is not obvious, although overall de-crease in these two parameters is a clear indication of enhancedcontribution from the Himalayan Rivers compared to March–Mayperiod. Additionally, the high salinity patch seen in the northernBOB in the previous season moves southwards, maintained possibly

Fig. 3. (a) The Sampling locations in the BOB from various cruises corresponding to period—March to May; (b) distribution of δ18O; and (c) surface water salinity. The acronyms are:IND—India; SL—Sri Lanka; Ch—Chennai; Vi—Visakhapatnam; Ko—Kolkata; BGD—Bangladesh; MM—Myanmar; Po—Port Blair. No smoothing function was applied while drawing thecontours in (b) and (c). The cruise names associated with cruise code is given in Table-1.



by the continued negative values of (P−E) (Fig. 2c; upper panel). Thecorresponding small patch of high δ18O in the northern BOB from theprevious season almost disappears except for minor remnants. Aprominent lowering of salinity and δ18O seen along the east coast ofIndia, off Visakhapatnam and further northeast is the result of mon-soonal river discharge from Mahanadi and Godavari and is possiblyshifted northwards along the coast in response to prevailing north-ward EICC (Fig. 2b; upper panel). The observed increase in δ18O andsalinity during this season in the southern BOB is possibly relatedto transfer of AS water via SMC around Sri Lanka (Fig. 2b; upperpanel). For the same reason, samples located south of ~10°N in theBOB also show high salinity (32–36; enclosed within G-1 in Fig. 6).

During winter months of November to February (Fig. 5), surfacewater with low values of salinity (b30) and δ18O (b−1‰) extendssouth of ~10°N, particularly on the east coast of India. This is interpretedas due to (i) influence of isotopically depleted river discharge from Pen-insular India during NE monsoon; (ii) positive (P−E) in the northernBOB (Fig. 2c; lower panel); and (iii) steady southward dispersal offresh water from Himalayan Rivers, advected by the prevailing south-ward moving EICC, along the east coast of India (Fig. 2b; lower panel).

The foregoing discussion indicates that the seasonal distribution ofδ18O and salinity over the northern BOB surface is governed domi-nantly by the spatial and temporal variations in (P+R− |E|) annually

amounting to ~4000 km3, aided by the prevailing EICC and seasonalvariation and differences in isotopic and ionic characteristic of Hima-layan and Peninsular Rivers. It is to be noted that (P+R− |E|) plays adominant role in governing the surface water characteristics of theBOB in spite of the fact that both SMC and WMC transfer much larger[>75 times (P+R− |E|)] amount of water (annually ~300,000 km3)between the AS and the BOB. This is possibly due to transport ofwater through SMC and WMC being distributed over ~400 m depth,mixing of which with upper most surface waters is restricted by theformation of shallow stratified plume of meteoric water over theBOB. However, during winter, in spite of low river discharge fromthe Peninsular Rivers, low salinity and low δ18O waters are persistentalong the east coast of India. This, as mentioned earlier, is possibly re-lated to the influx of isotopically depleted Peninsular River duringwinter monsoon and transport of Himalayan river discharges underthe influence of southward EICC, when the freshwater plume in theBOB is thinnest (Shetye et al., 1996). During summer, similar effectdue to reversal of EICC, in terms of high salinity and high δ18O valuesalong the east coast is not so prominent because all river dischargesare highest during this period and fresh water plume in BOB isthickest (Shetye et al., 1996).

Latitudinal variation in salinity and δ18O of all the 356 data pairs isplotted in Fig. 6. Outside the region dominated by the riverine influx

Fig. 4. Same as Fig. 3 for June to October.

Fig. 5. Same as Fig. 3 for November–February.



(i.e. south of ~10°N; enclosed in G-1) both salinity and δ18O are seento vary within a narrow range of high values (Salinity: 32 to 36; δ18O:0 to 0.5‰) regardless of the sampling season. Beyond this narrowrange, two groups (G-2 and G-3) have been identified between10°N and 22°N. Both these groups pertain to the regions that arestrongly affected by the riverine discharge. The group G-2 largely re-fers to the samples closer to the coasts that are influenced by freshwater influx derived either from winter (NE monsoon) precipitationfrom the east coast of India or the Himalayan river discharge outsidethe summer (SW) monsoon months. The samples in group G-2 showrelatively low values of both salinity (26–30) and δ18O (b−0.5‰) re-gardless of sampling season. The low δ18O values are the result of theriverine discharge generated during NE winter monsoon (Warrieret al., 2010) and/or the snow-melt component of the HimalayanRivers. The Himalayan riverine influx throughout the year has lowsalinity and low δ18O values (Lambs et al., 2005) but the PeninsularRivers have low salinity without correspondingly lower δ18O, exceptduring winter monsoon (Warrier et al., 2010). The group G-3 on theother hand represents samples all across the central BOB that are

Table 1Cruise names and codes as used for plotting the geographical distributions of salinityand δ18O in Figs. 3, 4 and 5.

Cruise names Cruise code Reference

Andaman to Visakhapatnam(Commercial Vessel)

1 This work

SK-63 2 Singh et al. (2010)SK-191 3 Singh et al. (2010)Chennai to Andaman and Back(ORV Sagarnidhi-1)

4 This work

Chennai to south of equator(ORV Sagarnidhi-2) 5 This workSK-182 6 Singh et al. (2010)G-200 7 Singh et al. (2010)Chennai to Andaman(Commercial Vessel)

8 This work

Andaman to Kolkata and back(Commercial Vessel)

9 This work

SK-70 10 Singh et al. (2010)SO-93 11 Singh et al. (2010)

Table 2Surface waters (depth 1–10 m) of the Bay of Bengal: sampling details and results of salinity and isotope measurements.

Sample ID Sampling date dd/mm/yy Cruise name Lat °N Long °E Salinity δ18O (‰) δD (‰) d (‰)

4827 01-12-07 Chennai to Andaman 12.74 82.94 30.20 −0.04 3.50 3.84828 01-12-07 Chennai to Andaman 12.74 83.01 29.50 0.00 5.02 5.14829 01-12-07 Chennai to Andaman 12.72 83.08 30.00 0.31 6.24 3.74830 01-12-07 Chennai to Andaman 12.71 83.19 29.90 −0.10 2.87 3.74831 01-12-07 Chennai to Andaman 12.69 82.25 30.30 −0.02 4.82 4.94832 01-12-07 Chennai to Andaman 12.69 83.30 30.40 −0.04 4.38 4.74833 01-12-07 Chennai to Andaman 12.65 83.40 30.40 −0.01 6.50 6.64834 01-12-07 Chennai to Andaman 12.63 83.62 31.10 −0.20 3.87 5.54835 01-12-07 Chennai to Andaman 12.61 83.73 31.00 −0.03 8.34 8.64836 01-12-07 Chennai to Andaman 12.59 83.82 31.30 −0.06 5.04 5.54837 01-12-07 Chennai to Andaman 12.57 83.77 31.10 −0.01 7.16 7.34838 01-12-07 Chennai to Andaman 12.57 83.90 30.80 0.00 5.29 5.34839 01-12-07 Chennai to Andaman 12.56 84.04 30.80 0.03 3.89 3.74840 01-12-07 Chennai to Andaman 12.54 84.13 30.20 0.01 4.43 4.44841 01-12-07 Chennai to Andaman 12.53 84.24 29.70 0.01 7.53 7.44842 01-12-07 Chennai to Andaman 12.52 84.35 29.70 −0.10 6.25 7.04843 01-12-07 Chennai to Andaman 12.50 84.46 29.40 −0.08 7.83 8.54844 01-12-07 Chennai to Andaman 12.47 84.57 29.00 −0.03 4.34 4.64845 01-12-07 Chennai to Andaman 12.45 84.68 28.00 0.03 5.52 5.34846 01-12-07 Chennai to Andaman 12.42 84.79 28.60 −0.62 5.12 10.14847 02-12-07 Chennai to Andaman 11.99 87.83 28.40 −0.03 3.55 3.84848 02-12-07 Chennai to Andaman 11.99 87.69 28.40 −0.03 7.82 8.14849 02-12-07 Chennai to Andaman 11.96 88.09 28.90 0.00 5.23 5.24850 02-12-07 Chennai to Andaman 11.95 88.17 29.40 −0.26 2.56 4.64851 02-12-07 Chennai to Andaman 11.93 88.36 29.00 0.06 7.02 6.54852 02-12-07 Chennai to Andaman 11.92 88.36 31.00 0.09 5.44 4.74853 02-12-07 Chennai to Andaman 11.90 88.46 31.00 0.05 5.97 5.54854 02-12-07 Chennai to Andaman 11.89 88.57 31.00 −0.17 4.89 6.24855 02-12-07 Chennai to Andaman 11.88 88.65 31.20 −0.05 4.83 5.24856 02-12-07 Chennai to Andaman 11.87 88.76 31.40 −0.30 3.25 5.74857 02-12-07 Chennai to Andaman 11.86 88.86 31.60 −0.10 4.20 5.04858 02-12-07 Chennai to Andaman 11.84 88.96 31.40 −0.11 4.16 5.04859 02-12-07 Chennai to Andaman 11.82 89.07 31.00 −0.04 5.33 5.74860 02-12-07 Chennai to Andaman 11.81 89.18 31.00 −0.04 4.83 5.14861 02-12-07 Chennai to Andaman 11.79 89.27 31.00 −0.07 2.73 3.34862 02-12-07 Chennai to Andaman 11.77 89.38 29.70 −0.03 3.73 4.04863 02-12-07 Chennai to Andaman 11.76 89.49 29.70 −0.02 2.70 2.94864 02-12-07 Chennai to Andaman 11.74 89.59 29.40 −0.14 3.29 4.44865 02-12-07 Chennai to Andaman 11.72 89.70 29.00 −0.28 2.37 4.64866 02-12-07 Chennai to Andaman 11.71 89.81 29.90 −0.24 2.07 4.04867 03-12-07 Chennai to Andaman 11.31 92.68 28.10 −0.31 1.57 4.14868 03-12-07 Chennai to Andaman 11.33 92.44 29.70 −0.52 2.36 6.54869 03-12-07 Chennai to Andaman 11.33 92.25 29.70 −0.31 0.96 3.44870 03-12-07 Chennai to Andaman 11.38 92.04 29.70 −0.25 0.88 2.84871 03-12-07 Chennai to Andaman 11.39 91.74 30.20 −0.28 2.56 4.84872 03-12-07 Chennai to Andaman 11.41 91.75 30.40 −0.33 3.01 5.64873 03-12-07 Chennai to Andaman 11.55 90.99 31.00 −0.28 4.24 6.54874 03-12-07 Chennai to Andaman – – 31.00 −0.25 1.35 3.43501 01-03-08 Andaman to Visakhapatnam 13.07 89.89 29.60 −0.23 −0.27 1.63502 01-03-08 Andaman to Visakhapatnam 13.22 89.71 29.70 −0.21 −0.98 0.73503 01-03-08 Andaman to Visakhapatnam 13.30 89.52 30.20 −0.24 0.65 2.6



Table 2 (continued)


3504 01-03-08 Andaman to Visakhapatnam 13.49 89.27 30.40 −0.50 0.59 4.63505 01-03-08 Andaman to Visakhapatnam 13.67 89.07 30.80 −1.97 −12.86 2.93506 01-03-08 Andaman to Visakhapatnam 13.77 88.94 31.00 −0.50 – –3507 01-03-08 Andaman to Visakhapatnam 13.89 88.72 31.00 −0.42 0.22 3.63508 01-03-08 Andaman to Visakhapatnam 14.02 88.53 31.00 −0.33 −0.52 2.13509 01-03-08 Andaman to Visakhapatnam 14.15 88.32 29.70 −0.36 1.64 4.53510 01-03-08 Andaman to Visakhapatnam 14.28 88.14 29.70 −0.10 0.83 1.63511 01-03-08 Andaman to Visakhapatnam 14.43 87.92 29.40 −0.05 0.25 0.73512 01-03-08 Andaman to Visakhapatnam 14.57 87.74 29.00 −0.07 4.90 5.43513 02-03-08 Andaman to Visakhapatnam 14.72 87.52 31.00 −0.51 3.56 7.63514 02-03-08 Andaman to Visakhapatnam 16.26 85.32 27.60 −0.07 7.47 8.13515 02-03-08 Andaman to Visakhapatnam 16.41 85.10 27.90 −0.07 5.70 6.33516 02-03-08 Andaman to Visakhapatnam 16.54 84.95 28.60 −0.50 3.87 7.83517 02-03-08 Andaman to Visakhapatnam 16.69 84.74 29.20 −0.07 3.00 3.63518 02-03-08 Andaman to Visakhapatnam 16.82 84.55 30.20 0.06 – –3519 02-03-08 Andaman to Visakhapatnam 16.94 84.37 30.40 −0.44 0.09 3.63520 02-03-08 Andaman to Visakhapatnam 17.07 84.19 31.00 −0.32 1.56 4.13521 02-03-08 Andaman to Visakhapatnam 17.22 84.00 31.00 −0.16 3.25 4.53522 02-03-08 Andaman to Visakhapatnam 17.35 83.82 31.00 −0.19 0.99 2.53523 02-03-08 Andaman to Visakhapatnam 17.48 83.63 31.20 −0.09 3.47 4.23524 02-03-08 Andaman to Visakhapatnam 17.60 83.43 30.00 −0.17 2.24 3.64941 01-06-08 Chennai to Andaman and back 13.46 82.71 30.80 0.11 4.83 3.94942 01-06-08 Chennai to Andaman and back 13.47 82.69 30.90 −0.07 3.68 4.24943 01-06-08 Chennai to Andaman and back 13.54 83.14 31.00 0.03 4.67 4.54944 01-06-08 Chennai to Andaman and back 13.52 83.36 30.80 −0.09 2.81 3.64945 01-06-08 Chennai to Andaman and back 13.52 83.36 30.90 0.03 4.63 4.44946 02-06-08 Chennai to Andaman and back 13.55 84.23 30.60 −0.21 2.51 4.24947 02-06-08 Chennai to Andaman and back 13.57 84.67 30.90 −0.06 3.80 4.34948 02-06-08 Chennai to Andaman and back 13.57 85.13 30.50 0.12 2.30 1.44949 02-06-08 Chennai to Andaman and back 13.59 85.67 30.10 −0.07 2.85 3.44950 02-06-08 Chennai to Andaman and back 13.60 86.17 30.30 −0.29 3.57 5.94951 03-06-08 Chennai to Andaman and back 13.65 88.10 29.50 −0.21 2.21 3.94952 03-06-08 Chennai to Andaman and back 13.67 88.44 29.90 −0.20 2.47 4.04953 03-06-08 Chennai to Andaman and back 13.68 89.06 29.90 −0.12 1.87 2.84954 03-06-08 Chennai to Andaman and back 13.70 89.29 30.30 −0.18 2.24 3.74955 03-06-08 Chennai to Andaman and back 13.72 89.85 29.20 −0.19 1.60 3.24957 04-06-08 Chennai to Andaman and back 13.77 92.19 29.00 −0.15 2.54 3.84958 04-06-08 Chennai to Andaman and back 13.79 92.70 29.30 −0.14 3.47 4.54959 04-06-08 Chennai to Andaman and back 13.79 92.90 29.30 −0.25 3.61 5.64875 05-06-08 Chennai to Andaman and back 12.29 93.84 29.30 −0.27 0.65 2.84876 05-06-08 Chennai to Andaman and back 12.29 93.88 29.30 −0.27 2.37 4.54877 05-06-08 Chennai to Andaman and back 12.29 93.88 29.30 −0.43 1.73 5.14878 05-06-08 Chennai to Andaman and back 12.27 93.89 29.30 −0.27 1.94 4.14879 05-06-08 Chennai to Andaman and back 12.26 93.88 29.30 −0.24 1.35 3.24880 05-06-08 Chennai to Andaman and back 12.26 93.86 29.30 −0.22 1.36 3.14881 05-06-08 Chennai to Andaman and back 12.26 93.85 29.30 −0.51 1.45 5.54882 05-06-08 Chennai to Andaman and back 12.29 93.84 29.30 −0.16 3.33 4.64883 05-06-08 Chennai to Andaman and back 12.34 93.22 29.30 −0.29 3.67 6.04960 05-06-08 Chennai to Andaman and back 13.73 93.81 28.80 −0.33 0.88 3.54961 05-06-08 Chennai to Andaman and back 12.34 93.82 28.70 −0.42 0.57 4.04962 05-06-08 Chennai to Andaman and back 12.31 93.82 30.00 −0.56 1.39 5.94963 05-06-08 Chennai to Andaman and back 12.34 93.22 28.80 −0.43 1.79 5.24964 06-06-08 Chennai to Andaman and back 12.44 92.85 29.60 −0.27 1.50 3.74965 06-06-08 Chennai to Andaman and back 12.34 92.75 30.20 −0.30 0.01 2.44966 06-06-08 Chennai to Andaman and back 12.35 91.72 29.40 −0.09 1.85 2.64967 06-06-08 Chennai to Andaman and back 12.33 92.70 30.20 −0.31 1.03 3.54968 06-06-08 Chennai to Andaman and back 12.30 92.12 29.40 −0.19 2.77 4.34969 07-06-08 Chennai to Andaman and back 12.69 89.64 29.70 −0.34 3.49 6.24970 07-06-08 Chennai to Andaman and back 12.76 89.18 31.00 −0.20 1.91 3.54971 07-06-08 Chennai to Andaman and back 12.84 88.16 30.90 −0.21 2.37 4.14972 08-06-08 Chennai to Andaman and back 12.36 85.35 31.00 −0.22 3.39 5.24973 08-06-08 Chennai to Andaman and back 12.44 84.68 31.60 −0.11 3.30 4.14974 08-06-08 Chennai to Andaman and back 12.55 84.09 31.60 −0.13 3.97 5.04975 09-06-08 Chennai to Andaman and back 13.01 81.17 31.70 0.00 3.65 3.64976 09-06-08 Chennai to Andaman and back 13.09 80.46 31.80 −0.03 2.81 3.04884 07-08-08 Chennai to south of equator 2.42 83.96 34.63 0.05 3.12 2.74888 10-08-08 Chennai to south of equator 0.00 79.01 34.53 0.08 – –4892 10-08-08 Chennai to south of equator −0.02 78.00 35.02 0.08 4.42 3.74893 12-08-08 Chennai to south of equator 2.50 80.49 34.47 0.18 5.58 4.14897 14-08-08 Chennai to south of equator 3.69 71.44 34.57 0.16 5.07 3.84901 15-08-08 Chennai to south of equator −0.78 80.48 34.62 0.05 3.94 3.64902 15-08-08 Chennai to south of equator −0.78 80.48 34.62 0.20 6.32 4.74906 18-08-08 Chennai to south of equator −2.50 80.48 34.59 0.06 5.49 5.04913 19-08-08 Chennai to south of equator −3.99 80.49 34.46 0.25 3.77 1.84918 19-08-08 Chennai to south of equator −3.99 80.49 34.48 0.28 7.14 4.94922 19-08-08 Chennai to south of equator −4.00 – 34.46 −0.03 2.17 2.4

(continued on next page)



affected by the variable extent of riverine influx during the SW sum-mer monsoon period.

All sample pairs are also plotted in Fig. 7 for ascertaining δ18O–S re-lationship. No significant relationship could be established between thetwo parameters considering all the samples derived from different

regions of BOB together. This is in spite of the fact that a broad geo-graphical correspondence between δ18O and salinity during differentseasons is seen (in Figs. 3, 4 and 5). This lack of δ18O–S relationshipcould possibly be due to (i) clumping of samples collected during differ-ent years/seasons; and (ii) absence of samples in this study from the

Table 2 (continued)


4923 19-08-08 Chennai to south of equator – – 34.46 0.06 2.53 2.04924 19-08-08 Chennai to south of equator – – 34.46 −0.34 0.86 3.64925 23-08-08 Chennai to south of equator 8.01 80.47 34.40 0.07 4.64 4.14926 14-09-08 Chennai to south of equator – – 34.50 −0.22 4.50 6.34927 14-09-08 Chennai to south of equator – – 34.99 0.02 3.87 3.74928 14-09-08 Chennai to south of equator – – 34.99 0.05 3.75 3.44929 14-09-08 Chennai to south of equator – – 34.90 0.06 4.96 4.54930 14-09-08 Chennai to south of equator – – 34.90 0.01 4.47 4.44931 14-09-08 Chennai to south of equator – – 35.20 0.05 4.97 4.64932 16-09-08 Chennai to south of equator – – 35.20 −0.21 3.86 5.54933 16-09-08 Chennai to south of equator – – 35.40 0.18 6.60 5.14934 16-09-08 Chennai to south of equator – – 35.50 0.06 5.78 5.34935 16-09-08 Chennai to south of equator – – 35.40 0.16 4.94 3.64936 16-09-08 Chennai to south of equator – – 35.40 1.00 7.73 −0.34937 16-09-08 Chennai to south of equator – – 34.80 0.15 5.16 3.94938 16-09-08 Chennai to south of equator – – 34.80 −0.13 4.09 5.14939 16-09-08 Chennai to south of equator – – 34.60 0.14 5.31 4.24940 16-09-08 Chennai to south of equator – – 34.60 0.15 5.56 4.33525 27-11-08 Andaman to Kolkata and back 11.91 93.36 30.20 −0.14 1.42 2.63526 27-11-08 Andaman to Kolkata and back 11.98 93.32 29.50 −0.08 1.23 1.93527 27-11-08 Andaman to Kolkata and back 11.97 93.25 30.00 −0.22 1.82 3.63528 27-11-08 Andaman to Kolkata and back 12.21 93.40 29.90 −0.21 0.53 2.23529 27-11-08 Andaman to Kolkata and back 12.18 93.23 30.30 −0.11 −0.99 −0.13530 28-11-08 Andaman to Kolkata and back 12.14 93.24 30.40 0.02 1.30 1.23531 28-11-08 Andaman to Kolkata and back 15.36 92.43 30.40 −0.20 0.42 2.03532 28-11-08 Andaman to Kolkata and back 15.45 92.17 31.10 −0.61 −1.14 3.73533 28-11-08 Andaman to Kolkata and back 15.39 92.23 31.00 −0.88 −3.66 3.43534 28-11-08 Andaman to Kolkata and back 15.55 92.08 31.30 −1.13 −6.03 3.03535 28-11-08 Andaman to Kolkata and back 16.09 91.74 31.10 −1.32 −9.34 1.23536 28-11-08 Andaman to Kolkata and back 16.16 91.67 30.80 −1.51 −8.64 3.43537 28-11-08 Andaman to Kolkata and back 16.16 91.65 30.80 −1.61 −8.19 4.73538 28-11-08 Andaman to Kolkata and back 16.36 91.67 30.20 −1.52 −8.67 3.53539 28-11-08 Andaman to Kolkata and back 16.27 91.70 29.70 −1.33 −5.51 5.13540 29-11-08 Andaman to Kolkata and back 19.37 89.39 29.70 −0.37 0.78 3.83541 29-11-08 Andaman to Kolkata and back 19.43 89.33 29.40 −0.28 1.28 3.53542 29-11-08 Andaman to Kolkata and back 19.43 89.39 29.00 −0.21 −0.58 1.13543 29-11-08 Andaman to Kolkata and back 19.47 89.23 28.00 −0.06 0.62 1.13544 29-11-08 Andaman to Kolkata and back 19.67 89.35 28.60 – – –3545 29-11-08 Andaman to Kolkata and back 19.68 89.34 28.40 −0.19 2.83 4.33546 29-11-08 Andaman to Kolkata and back 20.18 88.87 28.40 0.04 2.00 1.73547 29-11-08 Andaman to Kolkata and back 20.15 88.80 28.90 −0.22 2.24 4.03548 29-11-08 Andaman to Kolkata and back 20.32 88.73 29.40 −0.16 2.39 3.73549 29-11-08 Andaman to Kolkata and back 20.67 88.39 29.00 −0.18 1.67 3.13550 05-12-08 Andaman to Kolkata and back 19.63 89.13 30.20 −0.06 0.28 0.73551 05-12-08 Andaman to Kolkata and back 19.63 89.26 30.20 0.00 1.18 1.23552 05-12-08 Andaman to Kolkata and back 19.67 89.15 30.20 −0.14 0.06 1.23553 05-12-08 Andaman to Kolkata and back 19.53 89.29 30.20 −0.10 1.87 2.73554 05-12-08 Andaman to Kolkata and back 19.47 89.21 30.20 −0.07 1.43 2.03555 05-12-08 Andaman to Kolkata and back 19.55 89.34 30.20 −0.18 2.14 3.63556 05-12-08 Andaman to Kolkata and back 19.28 89.44 30.40 −0.03 2.34 2.63557 05-12-08 Andaman to Kolkata and back 18.88 89.79 31.10 −0.26 1.93 4.03558 05-12-08 Andaman to Kolkata and back 18.91 89.79 31.00 −0.27 0.92 3.13559 05-12-08 Andaman to Kolkata and back 18.76 89.71 31.30 −0.13 0.66 1.73560 05-12-08 Andaman to Kolkata and back 18.82 89.81 31.10 −0.49 0.89 4.83561 05-12-08 Andaman to Kolkata and back 18.68 89.85 30.80 −0.21 2.83 4.53562 05-12-08 Andaman to Kolkata and back 18.64 89.89 30.80 −0.13 0.21 1.23563 06-12-08 Andaman to Kolkata and back 16.19 91.52 30.20 −1.34 −8.37 2.43564 06-12-08 Andaman to Kolkata and back 16.15 91.70 29.70 −1.29 −7.05 3.33565 06-12-08 Andaman to Kolkata and back 15.96 91.81 29.70 −1.30 −8.74 1.63566 06-12-08 Andaman to Kolkata and back 15.59 91.95 29.40 −1.39 −7.90 3.23567 06-12-08 Andaman to Kolkata and back 15.60 91.93 29.00 −1.47 −7.03 4.83568 06-12-08 Andaman to Kolkata and back 15.64 92.14 28.00 −1.28 −4.43 5.83569 06-12-08 Andaman to Kolkata and back 15.19 92.50 28.60 −0.89 −3.67 3.43570 06-12-08 Andaman to Kolkata and back 15.00 92.42 28.40 −0.87 −3.72 3.33571 06-12-08 Andaman to Kolkata and back 14.89 92.67 28.40 −0.96 −7.88 −0.23572 06-12-08 Andaman to Kolkata and back 14.61 92.52 28.90 −0.80 −2.47 3.93573 06-12-08 Andaman to Kolkata and back 14.47 92.93 29.40 −0.71 −2.63 3.13574 06-12-08 Andaman to Kolkata and back 14.31 92.97 29.00 −0.65 −1.56 3.6



north and northeast BOB coastal region where the gradient is expectedto be the steepest. Samples with salinity >32 and δ18O values close to0‰ (Group G-1 in Figs. 6 and 7) are from 5°S to 10°N. Most of the

other samples with salinity b34 also have lower δ18O values (b0‰).These samples in Group G-2 (Fig. 6 and 7), as mentioned earlier, deriveeither from the region with riverine influx dominated by east flowingPeninsular Rivers of India duringwintermonsoon or the Himalayan riv-erine discharge outside the summer (SW) monsoon period. The groupG-3 with samples having salinity b32 and δ18O>−1% correspond tothe regions that are affected by the riverine influx during the SW sum-mer monsoon period. Samples in this group show a weak δ18O–S rela-tionship for the season June-October, suggesting mixing of watersfrom both Peninsular and Himalayan Rivers during the SW monsoon,with an end member like Group G-1 (with enriched values of δ18Oand salinity).

From the forgoing discussion it is obvious that, δ18O–S relation-ship in different parts of the BOB is a complex multi-end member sys-tem depending mainly on freshwater–seawater mixing processes andlocal (P−E) that vary with season and region. Since characteristics offreshwater sources in the BOB vary strongly between regions and sea-sons, in terms of δ18O, several seasonal and sub-regional relationshipshave been observed by earlier workers, as discussed in Singh et al.(2010). Therefore, pooling of data from different cruises, for thewhole of the BOB results only in a weak δ18O–S relationship for theJune–October season (Fig. 7).

In addition to the informationobtainable from δ18O to S relationship,the δ18O–δD relationships in oceanwater are also useful in studying theinfluence of precipitation and evaporation as both these parameters offresh water and seawater are similarly affected by fractionation andmixing processes.

A δ18O–δD scatter diagram for the BOB surface water samples col-lected from north of ~10°N is shown in Fig. 8. The regression line slope(8.5±0.4) for these samples is more than that for the available worldsurface ocean water data [δD=7.37(±0.17)×δ18O−0.72(±0.97);(Rohling, 2007; Schmidt et al., 1999)] but is similar, within statisticalerror, to the slope of the Global Meteoric Water Line [GMWL, δD=8.17(±0.17)×δ 18O−11.27(±0.5); (Rozanski et al., 1993)]. This isinterpreted as due to a very significant component of meteoric water(derived either by direct precipitation or through river runoff) in thenorthern BOB.

The slightly lower slope for world surface ocean water (upper250 m) has been ascribed to evaporation which preferentially en-riches the heavier isotopes (Rohling, 2007). Viewed in this context,the BOB surface waters north of ~10°N, are not in full conformitywith world surface ocean water line. This, as discussed previously, ispossibly due to a strong stratification of a shallow (10–20 m) surfacelayer over the BOB, formed by meteoric water, and the underlyingbarrier layer preventing its mixing with deeper layers (Prasanna

Fig. 6. Latitudinal variation in δ18O and salinity. The three groups of samplesrepresenting different geographical regions of the BOB are identified. Group G-1 repre-sents the region south of ~10°N where the effect of riverine influx is not significant.Group G-2 represents the region affected by fresh water from the Himalayan Riversor Peninsular Rivers in winter months. Group G-3 represents the region affected bythe riverine influx during SW summer monsoon.

Fig. 7. The δ18O–S plot of all the samples from the BOB. The three groups of samplesrepresenting different geographical region of the BOB as identified in Fig. 6 are markedin the three boxes. Note: The regression line shown is based on samples collected dur-ing June–October only.

Fig. 8. The δ18O–δD regression plot for all samples north of 10°N. Similar plot for datapairs south of 10°N is shown in inset-b. Inset-a shows similar plot for the data pairswithin –10°N to +10°N from the global ocean surface water data (Schmidt et al.,1999) and those from Srivastava et al. (2010, 2007).



Kumar et al., 2002). The intercept (4.2±0.2) of δ18O-δD regressionline for the BOB surface water north of 10°N is however, lower thanGMWL indicating evaporation even from the stratified layer.

In contrast, the BOB surface water samples collected from Chennai–Andaman (June, 2008)—a region just slightly north of ~10°N and thosefrom further south have δ18O–δD regression slopes (Fig. 8; inset b)lower (5.7±0.5) than the world surface ocean water line (7.37±0.17). This is possibly due to prevailing negative (P−E) for this regionof the BOB that receives no direct river influx, particularly during sum-mer. Since Inter Tropical Convergence Zone (ITCZ) moves northwardduring May–June (Menon, 2005; Rao, 1975), the region of negative(P−E) in the BOB is expected to encompass the Chennai–Andaman re-gion as well. To examine if similar effect is also observed in evaporationdominated surface ocean waters in other parts of the world, a subset ofthe global ocean water data (Schmidt et al., 1999; Srivastava et al.,2010, 2007) in the range−10°N to +10°N is plotted in Fig. 8 (inset-a).This subset gives a δ18O-δD regression slope of 5.5 (±0.7) which withinthe statistical error is comparable to the observed slope of 5.7 (±0.5) forthe samples from the evaporation dominated region south of ~10°N inthe BOB.

It is thus obvious that seasonal and geographical variation in thesalinity and δ18O of the surface waters of the BOB is influenced, in ad-dition to (P+R− |E|), by the differences in the isotopic and ioniccomposition of Himalayan and Peninsular river discharges and theirseasonal variation. With a view to investigate the impact of seasonalvariations in the isotopic and ionic characteristics of Himalayan andPeninsular river discharges, the δ18O–δD regression plot and time se-ries of δ18O, d-excess and electrical conductivity (EC) for two Himala-yan Rivers and five east flowing Peninsular Rivers draining into theBOB is shown in Fig. 9 (unpublished data from the IWIN NationalProgramme).

The Himalayan Rivers have δ18O–δD regression slope (7.14±0.21), similar to the world surface ocean water (7.37±0.17) withinstatistical uncertainty. Both these slopes are smaller than the regres-sion line slope (8.5±0.4) for the BOB surface waters north of ~10°N(sampled in March and November-December). The two HimalayanRivers have consistently low δ18O (b−6‰) and high d-excess values(>2‰) throughout the year, compared to the Peninsular Rivers.These Himalayan Rivers also show relatively small range of seasonalvariability in EC (100–400 μSi), δ18O (−6 to −12‰) and d-excess(+2 to +18‰).

In contrast to Himalayan Rivers, the Peninsular Rivers have δ18O–δDregression slope (5.31±0.35) smaller than the GMWL (~8), world sur-face ocean water (7.37±0.17) and the BOB surface waters north of~10°N (8.5±0.4). The regression slope for Peninsular Rivers, however,is comparable to that for the BOB surface waters south of ~10°N (5.7±0.05). A large range of seasonal variability is observed in the EC (400–1200 μSi), δ18O (+3 to −7‰) and d-excess (+20 to −40‰) in caseof Peninsular Rivers, particularly southern rivers (Cauvery, Krishnaand Pennar). Both the d-excess and δ18O values of the Peninsular Riversare much lower during October–December period (Fig. 9) when NEwintermonsoon is active and affects the east coastal states in the south-ern half of Peninsular India. This low δ18O-δD regression slope value forPeninsular Rivers indicates evaporation which is also indicated by theirlower d-excess values (≤0‰ almost throughout the year).

As mentioned earlier, the high value of δ18O–δD regression slope(8.5±0.4) for the BOB surface waters north of ~10°N, is greater thanthat for both Himalayan (7.14±0.21) and Peninsular (5.31±0.35) Riv-ers, suggesting strong influence of direct precipitation over the BOB.Though our sampling is limited to the months of March and wintermonsoon (Nov–Dec), this may be true for the summermonsoon seasonas well as most of the direct precipitation over the BOB occurs duringthis period as seen from Fig. 1b.

Thus, seasonal distribution of the isotopic and EC data of the Hima-layan and the Peninsular river discharge support the earlier drawn con-clusion that low δ18O and salinity values observed (Fig. 5b and c) along

east coast of India are strongly influenced by the river influx from win-ter discharge of Peninsular Rivers and the effect of the winter EICC thatbrings the Himalayan River discharge towards east coast of India.

On the other hand, relatively higher values of δ18O and salinity(Group G-1 in Fig. 6) and lower values of δ18O–δD regression slope(Fig. 8b) observed for samples south of ~10°N seem to derive their sig-natures from the discharge of the Peninsular Rivers that have relativelyhigher δ18O and salinity and lower d-excess, and negative values of(P+R− |E|), particularly during warmer months (March–October).

4. Summary and conclusion

The spatio-temporal variation in δ18O and salinity, in the surfacewater samples from the BOB, collected during 2007–2008 has been in-vestigated together with published data. A broad geographical corre-spondence between seasonal variation in spatial distribution of δ18Oand salinity is observed. In general, δ18O and salinity have lower valuesin north and northeast part of the BOB which progressively increase insouthwest direction in all the three seasons (March to May, June toOctober, and November to February). During winter months ofNovember to February, surface water with low values of salinity(b30) and δ18O (b−1‰) extends south of ~10°N, particularly on theeast coast of India. This is interpreted as due to influence of NEmonsoonriver discharge from Peninsular India together with steady southwarddispersal of freshwater fromHimalayan Rivers advected by the prevail-ing southward moving EICC.

In spite of the observed broad geographical correspondence be-tween δ18O and salinity during different seasons, the δ18O–S plot forall available data pairs considered together shows no significant rela-tionship between the two parameters though such relationships for dif-ferent seasons and sub-regions of the BOB have been observed byearlier workers. This is because fresh water influx from different riversmodifies the δ18O (and salinity) of the surface waters of the BOB differ-ently in different regions. In particular, the δ18O of Himalayan Riversand that resulting fromwinter precipitation in Peninsular India is signif-icantly lower than the riverine discharge from Peninsular Rivers duringSWsummermonsoon,whereas, in comparison to seawater, the salinityof all riverine sources is essentially similar. This difference in the spatialand temporal characteristics of riverine discharge in terms of δ18O, incontrast to near similar salinity characteristics, seems to be responsiblefor lack of strong δ18O–S relationship when the BOB as a whole isconsidered as a single unit. This may have important implications forpalaeoclimatic studies.

The δ18O–δD regression plot of samples north of ~10°N is similarin slope to the GMWL and is ascribed to a stratified nature of meteoricwater dominating the surface waters of the BOB, and the strong roleof direct precipitation. However for samples south of ~10°N, theδ18O–δD regression line has a significantly lower slope and compareswell with the subset of global oceans data between 10°N and 10°S andsmall discharge of Peninsular Rivers during summer monsoon. Thisimplies that world surface ocean water line is not applicable to re-gions which do not receive significant amount of meteoric watereither by direct precipitation or continental runoff.

In general variations in salinity and isotopes are related to the bal-ance between evaporation and precipitation but this study highlightsthat the seasonal distribution of δ18O and salinity over the northernBOB is dominantly governed by the spatial and temporal variation inthe (P+R− |E|; ~4000 km3a−1) in spite of the fact that both SMCand WMC transfer more than 75 times water between the AS and theBOB. This confirms the dominant role of a shallow stratified layer ofme-teoric water over the BOB in restricting vertical mixing. This aspectneeds to be considered when using δ18O–S relationship for thepalaeoclimatic reconstructions from sediment cores in the BOB.

Earlier studies have indicated that formation of this stratified me-teoric water layer is important to maintain the threshold SST over theBOB for charging the cloud-base air-mass with the required moist



static energy for sustenance of large scale deep convection in the at-mosphere during the summer monsoon. Significant holding back ofcontinental runoff in response to societal requirements or meltingaway of glaciers due to global warming therefore may limit the for-mation of this stratified meteoric water layer leading to possible un-desirable consequences for rainfall over south Asian region. Thisaspect needs to be further investigated.

Acknowledgments

The work reported here has been carried out under the aegis of aNational Programme on Isotope fingerprinting of Waters of India(IWIN), funded jointly by the Department of Science and Technology,Govt. of India vide grant no. IR/S4/ESF-05/2004, and the Physical Re-search Laboratory (PRL).

Fig. 9. The δ18O–δD regression plot and time series of δ18O, d-excess and electrical conductivity for two Himalayan Rivers (Ganga at Behrampur and Brahmaputra at Pandu) and fivePeninsular Rivers (Cauvery at Musiri, Godavari at Polavaram, Krishna at Wadenapalli, Mahanadi at Cuttack and Pennar at two stations of Chennur and Puspagiri) draining in to theBOB. This figure is based on the unpublished data from the IWIN National Programme. The two Himalayan Rivers have consistently low δ18O (b−6‰) and high d-excess values(>2‰) throughout the year, compared to Peninsular Rivers. These Himalayan Rivers also show relatively small range of seasonal variability in EC (100–400 μSi), δ18O (−6 to −12‰)and d-excess (+2 to+18‰). The Peninsular Rivers have distinctly lower slope and intercept of the δ18O–δD regression line, higher EC and lower d-excess compared to Himalayan Rivers.



References

Abe, O., et al., 2009. A 6.5-year continuous record of sea surface salinity and seawaterisotopic composition at Harbour of Ishigaki Island, southwest Japan. Isot. Environ.Heal. Stud. 45 (3), 247–258.

Benway, H.M., Mix, A.C., 2004. Oxygen isotopes, upper ocean salinity, and precipitationsources in the eastern tropical Pacific. Earth Planet. Sci. Lett. 224, 493–507.

Bigg, G.R., Rohling, E.J., 2000. An oxygen isotope data set for marine waters. J. Geophys.Res. 105 (C4), 8527–8535.

Biksham, G., Subramanian, V., 1980. Chemical and sedimentmass transfer in the Godavaririver basin in India. J. Hydrol. 46, 331–342.

Brassell, S.C., Eglinton, G., Marlowe, I.T., Pflaumann, U., Sarnthein, M., 1986. Molecularstratigraphy: a new tool for climatic assessment. Nature 320 (6058), 129–133.

Chakrapani, G.J., Subramanian, V., 1990. Factors controlling sediment discharge in theMahanadi River Basin, India. J. Hydrol. 117 (1–4), 169–185.

Clemens, S., Prell, W., Murray, D., Shimmield, G., Weedon, G., 1991. Forcing mecha-nisms of the Indian Ocean monsoon. Nature 353, 720–725.

Delaygue, G., Jouzel, J., Dutay, J.-C., 2000. Oxygen 18–salinity relationship simulated byan oceanic general circulation model. Earth Planet. Sci. Lett. 178 (1–2), 113–123.

Delaygue, G., Bard, E., Rollion, C., 2001. Oxygen isotope/salinity relationship in theNorthern Indian Ocean. J. Geophys. Res. 106 (C3), 4565–4574.

Deshpande, R.D., Gupta, S.K., 2008. National Programme on Isotope fingerprinting ofwaters of India (IWIN).

Deshpande, R.D., Gupta, S.K., 2009. IWIN National Programme for Isotopic Characteri-zation of Indian Hydrological Cycle: Preliminary Results. 11th ISMAS TriennialConference of Indian Society for Mass Spectrometry. Indian Society for Mass Spec-trometry, Hyderabad, pp. 95–107.

Deshpande, R.D., Gupta, S.K., 2012. Oxygen and hydrogen isotopes in hydrologicalcycle: new data from IWIN national programme. Proc. Ind. Natl. Sci. Acad. 78 (3),321–331.

Duplessy, J.C., 1982. Glacial to interglacial contrasts in the northern Indian Ocean. Na-ture 295 (5849), 494–498.

Duplessy, J.C., Bé, A.W.H., Blanc, P.L., 1981. Oxygen and carbon isotopic compositionand biogeographic distribution of planktonic foraminifera in the Indian Ocean.Palaeogeogr. Palaeoclimatol. Palaeoecol. 33 (1–3), 9–46.

Stable isotopes and ocean mixing. In: Ferronsky, V.I., Brezgunov, V.S. (Eds.), Handbookof environmental isotope geochemistry. : Mar. Environ. A, 3. Elsevier, Amsterdam(1–27 pp.).

Gadgil, S., 2003. The Indian monsoon and its variability. Annu. Rev. Earth Planet. Sci. 31,429–467.

Gadgil, S., Joseph, P.V., Joshi, N.V., 1984. Ocean–atmosphere coupling over monsoon re-gions. Nature 312 (5990), 141–143.

Kumar, B., et al., 2010. Isotopic characteristics of Indian precipitation. Water Resour.Res. 46, W12548. http://dx.doi.org/10.1029/2009WR008532.

Lambs, L., Balakrishna, K., Brunet, F., Probst, J.L., 2005. Oxygen and hydrogen isotopiccomposition of major Indian rivers: a first global assessment. Hydrol. Process. 19(17), 3345–3355.

River Inputs to Ocean Systems. In: Martin, J.M., Burton, J.D., Eisma, D. (Eds.), United Na-tions Press, Geneva, Switzerland (384 pp.).

Maurya, A.S., Shah, M., Deshpande, R.D., Gupta, S.K., 2009. Protocol for δ18O and δDanalyses of water sample using Delta V plus IRMS in CF mode with Gas Bench IIfor IWIN National Programme at PRL, Ahmedabad. In: Aggarwal, S.K., Jaison, P.G.,Sarkar, A., Kumar, P. (Eds.), 11th ISMAS Triennial Conference of Indian Society forMass Spectrometry, 314–317. Indian Society for Mass Spectrometry, Hyderabad,pp. 314–317.

Menon, P.A., 2005. Ways of Weather. National Book Trust, India, New Delhi 109 pp.Millero, F.J., 2011. Editorial for marine chemistry. Mar. Chem. 124 (1), 1.Millero, F.J., Feistel, R., Wright, D.G., McDougall, T.J., 2008. The composition of Standard

Seawater and the definition of the Reference-Composition Salinity Scale. Deep-SeaRes. I Oceanogr. Res. Pap. 55 (1), 50–72.

Prahl, F.G., Wakeham, S.G., 1987. Calibration of unsaturation patterns in long-chainketone compositions for palaeotemperature assessment. Nature 330, 367–369.

Prasad, T.G., 1997. Annual and seasonal mean buoyancy fluxes for the tropical IndianOcean. Curr. Sci. 73, 667–674.

Prasanna Kumar, S., et al., 2002. Why is the Bay of Bengal less productive during summermonsoon compared to the Arabian Sea? Geophys. Res. Lett. 29 (24), 2235.

Prell, W.L., Hutson, W.H., 1979. Zonal temperature-anomaly maps of Indian Oceansurface waters: Modern and Ice-Age patterns. Science 206, 454–456.

Ramesh, R., Subramanian, V., 1993. Geochemical characteristics of the major tropicalrivers of India, hydrology of warm humid regions. Proceedings of the YokohamaSymposium, July 1993: IAHS Publ., No. 216, pp. 157–164 (Yokohama).

Rao, K.L., 1975. India's Water Wealth. Orient Longman, New Delhi (225 pp.).Rao, R.R., Sivakumar, R., 2003. Seasonal variability of sea surface salinity and salt bud-

get of the mixed layer of the north Indian Ocean. J. Geophys. Res. 108 (C1), 3009.Rohling, E.J., 2007. Progress in paleosalinity: overview and presentation of new ap-

proach. Palaeogeography 22 (3), PA3215.Rostek, F., et al., 1993. Reconstructing sea surface temperature and salinity using δ18O

and alkenone records. Nature 364, 319–321.Rozanski, K., Araguas-Araguas, L., Gonfiantini, R., 1993. Isotopic patterns in modern

global precipitation, climate change in continental isotopic records. Am. Geophys.Union Monogr. 1–36.

Schmidt, G.A., Bigg, G.R., Rohling, E.J., 1999. Global Seawater Oxygen-18 Database- v1.20.http://data.giss.nasa.gov/o18data/.

Sengupta, D., Bharath Raj, G.N., Shenoi, S.S.C., 2006. Surface freshwater from Bay ofBengal runoff and Indonesian throughflow in the tropical Indian Ocean. Geophys.Res. Lett. 33 (22), L22609.

Shankar, D., Vinayachandran, P.N., Unnikrishnan, A.S., 2002. The monsoon currents inthe north Indian Ocean. Prog. Oceanogr. 52, 63–120.

Shenoi, S.S.C., Shankar, D., Shetye, S.R., 2002. Differences in heat budgets of the near-surface Arabian Sea and Bay of Bengal: implications for the summer monsoon.J. Geophys. Res. 107 (C6), 3052.

Shetye, S.R., et al., 1996. Hydrography and circulation in the western Bay of Bengal duringthe northeast monsoon. J. Geophys. Res. 101 (C6), 14011–14025.

Singh, A., Jani, R.A., Ramesh, R., 2010. Spatiotemporal variations of the δ18O–salinityrelation in the northern Indian Ocean. Deep-Sea Res. I 57, 1422–1431.

Somayajulu, B.L.K., Rengarajan, R., Jani, R.A., 2002. Geochemical cycling in the Hooghlyestuary, India. Mar. Chem. 79, 171–183.

Sprintall, J., Wijffels, S., Chereskin, T., Bray, N., 2002. The JADE and WOCE I10/IR6Throughflow sections in the southeast Indian Ocean. Part 2: velocity and trans-ports. Deep-Sea Res. II Top. Stud. Oceanogr. 49 (7-8), 1363–1389.

Sprintall, J., Wijffels, S.E., Molcard, R., Jaya, I., 2009. Direct estimates of the Indonesianthroughflow entering the Indian Ocean: 2004–2006. J. Geophys. Res. 114 (C7),C07001.

Srivastava, R., Ramesh, R., Prakash, S., Anilkumar, N., Sudhakar, M., 2007. Oxygen isotopeand salinity variations in the Indian sector of the Southern Ocean. Geophys. Res. Lett.34 (24), L24603.

Srivastava, R., Ramesh, R., Jani, R.A., Anilkumar, N., Sudhakar, M., 2010. Stable oxygen,hydrogen isotope ratios and salinity variations of the surface Southern IndianOcean waters. Curr. Sci. 99 (10), 1395–1399.

Strain, P.M., Tan, F.C., 1993. Seasonal evolution of oxygen isotope–salinity relationshipsin high-latitude surface waters. J. Geophys. Res. 98 (C8), 14589–14598.

Subramanian, V., 1993. Sediment load of Indian Rivers. Curr. Sci. 64, 928–930.Sud, Y.C., Walker, G.K., Lau, K., M, 1999. Mechanisms regulating sea-surface tempera-

tures and deep convection in the tropics. Geophys. Res. Lett. 26 (8), 1019–1022.UNESCO, 1981a. Background papers and supporting data on the practical salinity scale

1978.UNESCO, 1981b. The practical salinity scale 1978 and the international equation of

state of seawater 1980.Warrier, C.U., et al., 2010. Isotopic characterization of dual monsoon precipitation—

evidence from Kerala, India. Curr. Sci. 98 (11), 1487–1495.Wijffels, S., Sprintall, J., Fieux, M., Bray, N., 2002. The JADE and WOCE I10/IR6

Throughflow sections in the southeast Indian Ocean. Part 1: water mass distribu-tion and variability. Deep-Sea Res. II Top. Stud. Oceanogr. 49 (7-8), 1341–1362.

Yu, L., Weller, R.A., 2007. Objectively analyzed air-sea heat Fluxes for the global ice-freeoceans (1981–2005). Bull. Am. Meteorol. Soc. 88, 527–539.


Date post:	23-Oct-2020
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

(This is a sample cover image for this issue. The actual cover is … · 2019. 6. 24. · in Fig....

Documents