INTRODUCTION
Radiocarbon (14C) is primarily produced in the atmosphere by interaction between cosmic rays and atmospheric nitrogen. 14C in the atmosphere reacts with oxygen to form carbon dioxide and then enters the biosphere and hydrosphere. It reaches the ocean mainly through air–sea CO2 exchange process (Alves et al. Reference Alves, Macario, Ascough and Bronk Ramsey2018; Bhushan et al. Reference Bhushan, Somayajulu, Chakraborty and Krishnaswami2000; Dutta and Bhushan Reference Dutta and Bhushan2012). 14C concentration of dissolved inorganic carbon in surface seawater depends on 14C concentration of the atmosphere and oceanic subsurface waters. Since ocean subsurface waters remain isolated from the atmosphere for hundreds of years before shoaling up, its 14C concentration is generally lower than atmospheric 14C concentration. Combination of air–sea CO2 exchange, and upwelling reduces the 14C activity of the surface ocean reservoir as compared to that of the atmosphere, leading to reservoir effect (Stuiver and Polach Reference Stuiver and Polach1977; Alves et al. Reference Alves, Macario, Ascough and Bronk Ramsey2018). This causes an offset between 14C age of marine samples and corresponding atmospheric age, and this offset is called the reservoir age (R) (Stuiver and Braziunas Reference Stuiver and Braziunas1993). The varying intensity of air–sea CO2 exchange, upwelling and horizontal or lateral advection, results in different reservoir ages of surface ocean across the globe. Variation of regional R from the global average R value is called reservoir effect correction (∆R). Using marine calibration curve, the 14C age corresponding to calendar year of sample growth or collection provides the global average R value for that time. Mathematically, subtracting this global average R value from measured 14C age of marine sample yields ∆R (Stuiver and Braziunas Reference Stuiver and Braziunas1993; Reimer and Reimer Reference Reimer and Reimer2017; Alves et al. Reference Alves, Macario, Ascough and Bronk Ramsey2018). ∆R values are applied to 14C age of marine samples before calibration to correct for local reservoir effect. As reservoir age varies due to ocean circulation, upwelling and freshwater flux, apart from reservoir age correction it also helps in understanding the oceanography of the region. In the northern Indian Ocean, ∆R values observed for the Bay of Bengal region are lower compared to those for the Arabian Sea (Dutta et al. Reference Dutta, Bhushan and Somayajulu2001). Intense upwelling in the Arabian Sea during monsoon leads to higher ∆R values (Southon et al. Reference Southon, Kashgarian, Fontugne, Metivier and Yim2002), whereas relatively lower ∆R values in the Bay of Bengal (Dutta et al. Reference Dutta, Bhushan and Somayajulu2001) is due to highly stratified surface water which impedes vertical mixing (Thadathil et al. Reference Thadathil, Muraleedharan, Rao, Somayajulu, Reddy and Revichandran2007; Sijinkumar et al. Reference Sijinkumar, Clemens, Nath, Prell, Benshila and Lengaigne2016). Andaman basin situated on the eastern side of the Bay of Bengal receives large amount of freshwater from rivers.
Several scientific investigations in the Andaman region used 14C ages focusing on diverse subjects like past monsoonal variability (Rashid et al. Reference Rashid, Flower, Poore and Quinn2007; Achyutan et al. Reference Achyuthan, Nagasundaram, Gourlan, Eastoe, Ahmad and Padmakumari2014; Ali et al. Reference Ali, Hathorne, Frank, Gebregiorgis, Stattegger, Stumpf, Kutterolf, Johnson and Giosan2015; Ota et al. Reference Ota, Kawahata, Murayama, Inoue, Yokoyama, Miyairi, Aung, Hossain, Suzuki, Kitamura and Moe2017; Kumar et al. Reference Kumar, Band, Ramesh and Awasthi2018; Bhushan et al. Reference Bhushan, Yadava, Shah, Banerji, Raj, Shah and Dabhi2019b), past salinity changes (Sijinkumar et al. Reference Sijinkumar, Clemens, Nath, Prell, Benshila and Lengaigne2016), deformational history of Andaman Islands (Rajendran et al. Reference Rajendran, Rajendran, Earnest, Prasad, Dutta, Ray and Anu2008; Kunz et al. Reference Kunz, Frechen, Ramesh and Urban2010; Awasthi et al. Reference Awasthi, Ray, Laskar and Yadava2013), past volcanic activity (Awasthi et al. Reference Awasthi, Ray, Laskar, Kumar, Sudhakar, Bhutani, Sheth and Yadava2010), past sea-level changes (Scheffers et al. Reference Scheffers, Brill, Kelletat, Brückner, Scheffers and Fox2012), past tsunami deposits (Jankaew et al. Reference Jankaew, Atwater, Sawai, Choowong, Charoentitirat, Martin and Prendergast2008) and archeological history (Cooper Reference Cooper1993) of the region. However, there are limited reservoir age estimates available from the region (Dutta et al. Reference Dutta, Bhushan and Somayajulu2001; Southon et al. Reference Southon, Kashgarian, Fontugne, Metivier and Yim2002). In order to constrain the reservoir effect and understand the oceanography of the region, more pre-bomb 14C values from this region are required. Corals are good marine archive recording the past 14C changes in DIC of seawater (Druffel and Linick Reference Druffel and Linick1978; Hideshima et al. Reference Hideshima, Matsumoto, Abe and Kitagawa2001; Grumet et al. Reference Grumet, Guilderson and Dunbar2002; Dang et al. Reference Dang, Mitsuguchi, Kitagawa, Shibata and Kobayashi2004; Druffel et al. Reference Druffel, Robinson, Griffin, Halley, Southon and Adkins2008). In this study, annually banded Porites coral core drilled from the Landfall Island from the northern Andaman has been analyzed for its 14C composition. The pre-bomb 14C value between 1948 and 1951 obtained from the coral has been used to estimate the reservoir age correction of the Andaman basin.
MATERIALS AND METHODS
In March 2018, a 126-cm-long coral core was collected from a live Porites sp. colony from Landfall Island (13º39′N, 93º02′E) situated in the northern Andaman using an underwater coral driller. The coral core was cut into 8-mm-thick slices. X-radiograph of the coral slice shows annual density banding (Figure 1). The coral core slices were treated with 10% H2O2 solution and then cleaned with Milli-Q in ultrasonic bath to remove organics from coral skeleton. After cleaning, the slices were dried at 60ºC. The clean and dried slices were drilled for stable isotope and 14C analysis using micro-driller. Chronology of the annual bands were assigned using stable isotopic composition of the skeleton. The δ18O and δ13C values of drilled coral sample show good seasonality, which is also observed in another Porites coral core from Andaman region (Rixen et al. Reference Rixen, Ramachandran, Lehnhoff, Dasbach, Gaye, Urban, Ramachandran and Ittekkot2011). The distance between consecutive maxima of δ18O was considered as one year. Top most band was assigned year of sample collection i.e. 2018, and the last analysed band corresponds to year 1948. About 10 mg of drilled coral carbonate powder samples along with new oxalic acid standard (NIST Oxalic Acid, SRM 4990C, HOxII), inter-comparison sample VIRI-R (Scott et al. Reference Scott, Cook and Naysmith2010) and in-house coral standard (PRL-C) were converted to graphite using automated graphitization equipment (AGE3) (Wacker et al. Reference Wacker, Němec and Bourquin2010, Reference Wacker, Fülöp, Hajdas, Molnár and Rethemeyer2013). Oxalic acid standard (HOxII) is the primary standard used for normalizing sample 14C content, whereas VIRI-R was used as a check standard. The graphitized samples and standards were pressed into targets and measured at PRL Accelerator Mass Spectrometer facility (PRL-AURiS; Bhushan et al. Reference Bhushan, Yadava, Shah and Raj2019a, Reference Bhushan, Yadava, Shah, Banerji, Raj, Shah and Dabhi2019b). Each target was measured for 10 cycles of 10,000 14C counts, with total of at least 100,000 14C counts. VIRI-R standard yielded 14C value of 74.6 ± 1.0 pMC, which is close to the consensus value of 73.338 ± 0.037 pMC, affirming the accuracy of the measurement. To determine the reservoir age of the region where the coral was growing, 14C concentration of six samples corresponding to period between 1948 and 1951 were analyzed.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20201109095843279-0655:S0033822220000910:S0033822220000910_fig1.png?pub-status=live)
Figure 1 X-radiograph of Landfall coral showing annual density banding along with its δ18O composition. White box is the marked area of samples between 1948 and 1951.
RESULTS AND DISCUSSION
Results of 14C measurement are summarized in Table 1, where ∆14C (‰) and 14C age (yr BP) were calculated using measured 14C/12C and 13C/12C ratios. The results have been reported following conventions of Stuiver and Polach (Reference Stuiver and Polach1977). Calculated ∆14C are corrected for fractionation and age between year of measurement and growth of coral band. As samples belong to 20th century, Suess correction of –9 ± 3‰ is applied only to ∆14C values of corals (Southon et al. Reference Southon, Kashgarian, Fontugne, Metivier and Yim2002). Model 14C age is derived from Marine13 calibration curve, which uses IntCal13 curve and ocean–atmosphere box diffusion model to obtain global marine surface ocean curve for 0 to 10.5 cal kBP (Reimer et al. Reference Reimer, Bard, Bayliss, Beck, Blackwell, Bronk Ramsey, Buck, Cheng, Edwards, Friedrich, Grootes, Guilderson, Haflidason, Hajdas, Hatté, Heaton, Hoffmann, Hogg, Hughen, Kaiser, Kromer, Manning, Niu, Reimer, Richards, Scott, Southon, Staff, Turney and van der Plicht2013). The model 14C age used here for period 1948–1951 is 469 ± 23 yr BP. ∆R values are calculated by subtracting model 14C age from conventional 14C age of the coral samples. Dutta et al. (Reference Dutta, Bhushan and Somayajulu2001) and Southon et al. (Reference Southon, Kashgarian, Fontugne, Metivier and Yim2002) did not use Suess corrected ∆14C value to calculate the ΔR values. Although, Southon et al. (Reference Southon, Kashgarian, Fontugne, Metivier and Yim2002) reported Suess corrected ∆14C values, but the ΔR values were calculated without Suess correction. Thus, to maintain uniformity and ease for comparison, ΔR values were calculated without Suess correction.
Table 1 Results of 14C analysis of Landfall coral skeleton between 1948–1951.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20201109095843279-0655:S0033822220000910:S0033822220000910_tab1.png?pub-status=live)
Errors quoted for ∆14C, 14C age and ∆R are one sigma. The ∆14C value for Landfall coral sample ranges from –32 ‰ to –54‰ with mean value of –40 ± 7‰ (mean ± SD, n = 6) between 1948 and 1951. The Suess-corrected ∆14C value averages around –31 ± 7‰. Between 1948 and 1951, ∆R value recorded by the Landfall coral ranges between –23 to –206 yr. The χ2 test was carried out to test the variability in our coral ΔR values. It is observed that χ2/(n–1) is less than 1, suggesting that the measurement errors explain the coral ΔR variability and no additional uncertainty is required when calculating the average ΔR (Mangerud et al. Reference Mangerud, Bondevik, Gulliksen, Hufthammer and Høisæter2006). The observed changes in 14C values of coral could result from local oceanographic conditions. Reservoir correction, calculated using Marine13-derived model age, for Chilika lake in the northern Bay of Bengal (Dutta et al. Reference Dutta, Bhushan and Somayajulu2001) equals to –61 ± 61 yr. As it is the only reservoir correction value available from the region, it is assumed to be representative of the northern Bay of Bengal. This value is lower than observed ∆R value of –23 ± 76 yr in the year 1950 for the Landfall coral, suggesting lateral mixing of surface waters from the northern Bay of Bengal may not result in such change in ∆R value. This suggests that the observed variation in coral 14C values could have resulted from either vertical mixing or lateral transport from the southern Andaman region. The mean ∆R value of Landfall coral is calculated to be –138 ± 61 yr. In absence of any other ∆R value reported from the northern Andaman, the obtained mean value of –138 ± 61 yr can be applied on 14C dates for the northern Andaman region for reservoir correction. Previously, Dutta et al. (Reference Dutta, Bhushan and Somayajulu2001) and Southon et al. (Reference Southon, Kashgarian, Fontugne, Metivier and Yim2002) had reported 14C values of bivalve (Asaphis deflavata) and gastropod (Thais sp.) shell from the Andaman region. Bivalve shell from the Stewart Sound in the northern Andaman gave ∆14C value of –55 ± 4‰ (Dutta et al. Reference Dutta, Bhushan and Somayajulu2001), and gastropod shell from the Nicobar Island showed ∆14C value of –53.6 ± 7.7‰ (Southon et al. Reference Southon, Kashgarian, Fontugne, Metivier and Yim2002). The ∆14C values from Landfall coral are higher when compared to the Stewart Sound and Nicobar Island samples.
The ∆R values calculated from the reported ∆14C of calcareous shells from the Stewart Sound and Nicobar Island is 12 and 30 yr, respectively (Table 2). Both these values are higher than ∆R value of the Landfall coral. It is interesting to note that even the highest ∆R value recorded by Landfall coral is lower than that of Andaman mollusk shells. The model ages used for reservoir age calculation for each sample are different as the year of growth or collection for these samples varies from 1913 to 1951. The model age ranges from 448 to 469 yr BP. The observed differences in ∆R value of samples are result of either species specific 14C activity or oceanic processes like upwelling and circulation or both.
Table 2 Reservoir age correction (ΔR) values of pre-bomb marine samples from Andaman Basin.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20201109095843279-0655:S0033822220000910:S0033822220000910_tab2.png?pub-status=live)
Feeding habits and habitats of mollusks can have effects on their 14C records. Species-dependent 14C activity can result in variable ∆R values for the same region (Dye Reference Dye1994; Forman and Polyak Reference Forman and Polyak1997; Hogg et al. Reference Hogg, Higham and Dahm1997; Petchey et al. Reference Petchey, Ulm, David, McNiven, Asmussen, Tomkins, Richards, Rowe, Leavesley, Mandui and Stanisic2012). Bivalves can be suspension feeders or deposit feeders. Some bivalves engage in deposit feeding depending on their local conditions (Petchey et al. Reference Petchey, Phelan and White2004). Thus, bivalves feeding on detritus of old limestone can result in high 14C ages. Petchey et al. (Reference Petchey, Phelan and White2004) analyzed marine shells from the Coral Sea and the Solomon Sea region, and they found mollusk (Asaphis violascens) collected from an area dominated by calcareous bedrock yielded high ∆R values. Unlike bivalves, Thais sp. is a carnivorous predator. The 14C content of these gastropod may not represent seawater DIC 14C content, as their 14C content depends on the carbon reservoir of their prey (Hua Reference Hua, Rink and Thompson2015). Lindauer et al. (Reference Lindauer, Santos, Steinhof, Yousif, Phillips, Jasim, Uerpmann and Hinderer2017) had also observed influence of food resource and habitat on the ∆R of bivalve and gastropod from Gulf of Oman region. Therefore, enriched values in mollusk (Asaphis deflavata, Thais sp.) from Andaman could be due to species specific 14C activity. Apart from species difference, the sample location can also be one of the reasons behind observed differences in ∆R value from Andaman region.
The Landfall Island is in the northern part of Andaman archipelago, which receives large flux of fresh riverine water. Salinity in the Andaman basin increases southwards (Babu and Sastry Reference Babu and Sastry1976) and reservoir age correction also shows increasing value with locations further south in the Andaman basin away from freshwater region. During winters, southern Andaman sea is influenced by flows from the Malacca strait originating from the South China Sea (Raju et al. Reference Raju, Gouveia and Murty1981). The average ∆R value reported for the South China Sea is –3 ± 50 yr (Dang et al. Reference Dang, Mitsuguchi, Kitagawa, Shibata and Kobayashi2004). During summer monsoon, Southwest Monsoon Current flows eastward south of Sri Lanka to bring saltier Arabian Sea waters into the Bay of Bengal (Schott et al. Reference Schott, Xie and McCreary2009). Southon et al. (Reference Southon, Kashgarian, Fontugne, Metivier and Yim2002) reported ΔR values of 127 yr for the Sri Lankan region. During the same period (summer), southern Andaman (around 10-degree channel) receives strong influx of surface currents from the Bay of Bengal side (Kiran Reference Kiran2017), which could bring ΔR enriched waters to the southern Andaman. These surface currents in summer and winter season could lead to the observed high ΔR values in the southern Andaman. Whereas, the northern Andaman region receives freshwater flux from rivers leading to stratification of surface waters, which inhibits vertical mixing and contributes to the lower reservoir age of the region. By comparing previously reported ΔR values with Landfall coral value, it is observed that there exists significantly large variation in ∆R values from the Andaman Sea derived from coral and mollusk shells (Dutta et al. Reference Dutta, Bhushan and Somayajulu2001; Southon et al. Reference Southon, Kashgarian, Fontugne, Metivier and Yim2002) (Figure 2). Reservoir age correction values from the northern Andaman Sea and the Bay of Bengal are lower as compared to the southern Andaman Sea. These variations in reservoir age corrections need to be accounted while correcting 14C dates of marine samples for reservoir age of the region.
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20201109095843279-0655:S0033822220000910:S0033822220000910_fig2.png?pub-status=live)
Figure 2 Map of northeastern Indian Ocean with reservoir age correction values from Landfall Island (this study), Chilika lake, Stewart sound (Dutta et al. Reference Dutta, Bhushan and Somayajulu2001) and Nicobar Island (Southon et al. Reference Southon, Kashgarian, Fontugne, Metivier and Yim2002). (Inset: study location marked by rectangle in northern Indian Ocean.)
CONCLUSION
A Porites coral core from Landfall Island in the northern Andaman basin was analyzed for its 14C concentrations. The ∆14C values of coral for the period 1948–1951 varies between –32 to –54‰. The mean ∆R value obtained for this period from the Landfall coral is –138 ± 61 yr, which is lowest reported for the northern Indian Ocean. The ∆R values reported from the Andaman basin shows large variations, wherein southern Andaman ∆R value is higher than that of the northern Andaman and Bay of Bengal. As the northern Andaman basin receives more freshwater flux as compared to the southern Andaman, such differences in reservoir age could be observed. However, difference in ∆R values due to species dependent 14C variability cannot be ruled out. More pre-bomb samples need to be analyzed to better estimate reservoir age and its variation with time in the Andaman basin.
ACKNOWLEDGMENT
We are extremely thankful to the Ministry of Earth Sciences (MoES) for funding the GEOTRACES project under which this work was carried out. We are extremely grateful to Director, PRL for his support. We thank the Ministry of Environment and Forest (MoEF) for granting permission for sampling of corals. We are grateful to PRL workshop and in particular to Rajesh Kaila for his help and support in the field and laboratory. We are thankful to Prof. PM Mohan of Pondicherry University, Port Blair, for his guidance and local logistic support during fieldwork.