Revisiting the Biostratigraphic Range and Possible Cause of the First and Last Occurrence of Globigerinoides Ruber (Pink) in the Northern Indian Ocean

Establishing the chronology of sediments is the most crucial aspect of paleoclimate reconstruction. Several dating methods are used to obtain the relative and absolute age of sediments. The use of biostratigraphic markers is one of the most classic and straightforward methods to assign dates to samples (Bown, 1996; de Kaenel et al., 1996; Ferreira et al., 2019). The First Occurrence Datum (FO) and the Last Occurrence Datum (LO) of different organisms are used as biostratigraphic markers (Macchioni & Cecca, 2002; Portilho-Ramos et al., 2014; Saraswat et al., 2019). One such biostratigraphic marker, the LAD of Globigerinoides ruber (pink) (120 ka), has been extensively used in the Indian and Pacific Ocean sediments (Thompson et al., 1979).

Globigerinoides ruber (pink) is a tropical to subtropical species (Schiebel & Hemleben, 2017), calcifying in the shallowest part of the mixed layer (Hemleben et al., 1989; Jonkers & Kučera, 2017). It is a warm water species preferring higher temperatures as compared to G. ruber (white). Therefore, G. ruber (pink) proliferates only in the summer and is often absent during the winter (Bé, 1960; Cifelli, 1965; Tolderlund & Bé, 1971; Williams et al., 1981; Hilbrecht, 1996; Kemle-von-Mücke & Hemleben, 1999; Richey et al., 2019). The optimum abundance of G. ruber (pink) is during the summer/fall season (Wilke et al., 2009; Haarmann et al., 2011). The relative abundance of G. ruber (pink) is exclusively coupled with the thermal maximum (Williams et al., 1981). It is considered as one of the species with an ecological affinity for the warmest water in the world ocean (Niebler et al., 1999).

The pink colour in G. ruber shells comes from a pigment called pheophytin, present in the shell lamellae (Bé & Hamlin, 1967). The pigment belongs to the zooxanthellae, symbiotically present in different chambers of G. ruber (Lee & McEnery, 1983). Globigerinoides ruber (pink) responds very quickly to hydrographic changes (Deuser & Ross, 1989). It is a dominant taxon in modern sediments of the Atlantic Ocean and Mediterranean Sea. However, it went extinct from the Indian and the Pacific Oceans around 120 ka (Thompson et al., 1979). Although the G. ruber (pink) LO is often used as a biostratigraphic marker, its FO and the possible cause of its first and last occurrence from the Indian Ocean have not been studied in detail. Therefore, documenting its first and last occurrence will not only provide an excellent biostratigraphic marker but will also help researchers understand abrupt changes in the global climate.

The main objective of this work is to constrain the FO and LO of G. ruber (pink) in the northern Indian Ocean and to explore the reasons for its first and last occurrence.

The Bay of Bengal receives a huge amount of freshwater and sediment flux from many rivers (Ganga, Brahmaputra, Mahanadi, Godavari, and Cauvery Rivers; Milliman & Meade, 1983; Milliman & Syvitski, 1992). Additionally, the water from the western Pacific Ocean also makes its way into the Bay of Bengal via the Indonesian Throughflow (Rahman et al., 2021). Both the Arabian Sea and the Bay of Bengal are affected by the seasonal reversal of monsoon winds. The monsoon winds change the direction of the East India Coastal Current (EICC), flowing southward between January and August and poleward from September to December (Varkey et al., 1996; Shetye et al., 1991, 1996) in the Bay of Bengal. This creates coastal upwelling during the summer (Sasamal et al., 2005) and strong stratification during the monsoon and post-monsoon season (Sarma et al., 2013, 2018). The upwelling brings nutrient-rich water to the surface, which increases productivity (Sarma et al., 2020). The baroclinic instability in the EICC forms mesoscale eddies, which increases the productivity in the western Bay of Bengal (Kurien et al., 2010; Chen et al., 2018; Cheng et al., 2018). The organic matter flux creates an oxygen deficient zone (ODZ) during both the monsoon and post-monsoon seasons (Sarma et al., 2013, 2016).

The riverine influx in the Arabian Sea is mainly confined to the north-eastern region by the Indus and small rivers draining from the Western Ghats of India (Milliman & Syvitski, 1992). Similar to the EICC in the Bay of Bengal, the West India Coastal Current (WICC) also reverses direction during summer and winter monsoon (Shankar et al., 2002). The summer monsoon current creates strong upwelling on the western margin of the Arabian Sea (Chatterjee et al., 2019). The high productivity during both the summer and winter seasons, coupled with restricted ventilation and fresh water capping, creates a persistent ODZ at the intermediate depths of the Arabian Sea (Naqvi & Naronha, 1991).

As a result of the difference in the riverine influx, the surface salinity is higher in the Arabian Sea as compared to the Bay of Bengal (Prasanna Kumar et al., 2002). The overall ODZ intensity in the Bay of Bengal is weaker compared to the Arabian Sea. The absence of strong upwelling due to salinity stratification and incomplete bacterial oxidation in the water column due to the rapid sinking of organic matter is the reason for the weaker ODZ in the Bay of Bengal as compared to other basins (Ittekkot et al., 1991; Rao et al., 1994; Naqvi et al., 1994; McCreary et al., 2013).

We used foraminiferal assemblage data from high-resolution samples available from recently drilled sediments of the northern Indian Ocean: sediments drilled in the western Bay of Bengal and the eastern Arabian Sea during the International Ocean Discovery Program Expeditions 353 (Site U1446) and 355 (Site U1457). In the Bay of Bengal, Site U1446 is located at a water depth of 1430 m, ∼75 km from the eastern margin of India, in the Mahanadi offshore basin (19°5.02′N, 85°44.08′E). In the Arabian Sea, Site U1457 is located at a water depth of 3534 m, on the western edge of the Lakshmi Basin (17°9.95’N, 67°55.80’E; Fig. 1).

A total of 328 samples from Site U1446 and 274 samples from Site U1457 were used (Appendices 1 and 2). The relative abundance data of G. ruber (pink), for the last 320 kyr, at Site U1457 was taken from Saraswat et al. (2019). Additional samples from Site U1457 to cover the time frame until 584 ka were processed following the standard procedure for processing sediments for foraminiferal studies (Saalim et al., 2017). A small aliquot of the sample was freeze-dried, weighed and wet sieved using a 63-µm sieve. The material retained on the sieve (coarse fraction, CF) was dried, weighed, and transferred to storage vials. A part of the CF was dry sieved at 125 µm. A weighed aliquot of CF>125 µm was used for picking. A minimum of 300 planktic foraminifera were picked and mounted on a micropaleontological slide. The entire >125-µm CF was used when 300 planktic foraminiferal shells were not available. The number of individuals of different planktic foraminifera species was counted. The species were identified using the plates published by Parker (1962) and Schiebel & Hemleben (2017). The samples with less than 100 total specimens were excluded from this study. The species nomenclature follows the recent taxonomic work of Brummer & Kučera (2022). However, instead of G. ruber ruber we have used G. ruber (pink) for the ease of comparison with previously published biostratigraphic studies.

Globigerinoides ruber (pink), although rare in the northern Indian Ocean, its LO has often been used as a biostratigraphic marker. The precise timing and cause of its FO are, however, debated. The relative abundance of G. ruber (white) doesn't show any relation with that of G. ruber (pink), indicating that the first and last occurrence of the latter are unrelated to that of the former, and both species have different ecological preferences.

Interestingly, Bhattacharjee et al. (2013) purportedly found G. ruber (pink) in the surface sediments at two stations in the western Bay of Bengal. However, none of the other previous surface distribution and sediment trap studies of planktic foraminifera in the northern Indian Ocean reported any presence of G. ruber (pink) in recent sediments (e.g., Duplessy et al., 1981; Cullen & Prell, 1984; Zhang, 1985; Naidu, 1993; Guptha et al., 1997; Peeters et al., 2002; Stoll et al., 2007; Bhadra & Saraswat, 2021; Maeda et al., 2022). In our recently published results (Bhadra & Saraswat, 2021), we collected surface sediment samples from the entire eastern margin of India, which includes stations that are near the study area of Bhattacharjee et al. (2013). We also didn't find any G. ruber (pink) in the surface sediments. Therefore, we have not considered the purported findings of G. ruber (pink) in the surface sediments of the northern Indian Ocean by Bhattacharjee et al. (2013) further in our manuscript.

We report the last occurrence of G. ruber (pink) at IODP Site U1446 at 128 ka and at Site U1457 at 123 ka, thus marking the LO in the Bay of Bengal and the Arabian Sea, respectively. The LO of G. ruber (pink) in the Indian and the Pacific Ocean is often used as a biostratigraphic marker to represent sediments older than 120 ka (Thompson et al., 1979). Additionally, we provide the FO of G. ruber (pink) at 399 ka in the Bay of Bengal. We propose that the complete absence in older sediments and the sudden occurrence of G. ruber (pink) at 399 ka can also be used as a biostratigraphic marker in the Bay of Bengal. However, as the chronology of both the sites has been obtained by comparing the oxygen isotopic ratio of foraminifera with the LR04 global isostack, there is an age uncertainty of 4 kyr for ages estimated for the last 1 Myr (Lisiecki & Raymo, 2005). Globigerinoides ruber (pink) appears at 568 kyr at Site U1457. But, if we exclude the minor sporadic intervals where only few specimens were found, the acme can be seen at MIS 11 in the eastern Arabian Sea, as well. Therefore, the occurrence of G. ruber (pink) only between MIS 11 and MIS 5e and its absence in both the older and younger sediments of the northern Indian Ocean is intriguing. The same has been reported in other parts of the Indian Ocean, as well. The first occurrence of G. ruber (pink) in the western Bay of Bengal matches well with that reported from the South China Sea (at 400 kyr; Li, 1997). Tian et al. (2002) also placed the FO of G. ruber (pink) in MIS 11 (at 407 kyr). A similar FO and LO of G. ruber (pink) during MIS and 10/11 MIS 5e, respectively, were reported in the southeastern Indian Ocean (Wells & Chivas, 1994). But, no attempt has been made to explore the reason behind the first and last occurrence of G. ruber (pink).

Prior to MIS 11, Globigerinoides ruber (pink) was completely absent in the Bay of Bengal. However, in the Arabian Sea, a few specimens of G. ruber (pink) occur prior to MIS11. Interestingly, the presence of G. ruber (pink) prior to MIS11 was not only rare in the Arabian Sea, but also with very faint colouration. Also, there is a 5-kyr difference in the last occurrence of G. ruber (pink) between both the basins. Globigerinoides ruber (pink) appeared late in the Bay of Bengal and disappeared early. It should, however, be noted here that the oxygen isotopic-ratio-based chronology has a large error associated with it.

The extreme environmental preference (warmest temperature and low salinity), as well as the sensitivity of the associated symbionts, makes G. ruber (pink) vulnerable to fluctuation in ambient conditions. Therefore, we explore the potential impact of changing environmental conditions on the first and last occurrence of G. ruber (pink). The interval prior to the first occurrence of G. ruber (pink) in the Bay of Bengal, namely MIS 12, was an extremely cold period due to lower insolation (Raymo, 1997; Paillard, 2001). The Asian monsoon collapsed during MIS 12 (Barth et al., 2018). Subsequently, there was a sudden increase in the CO₂ concentration after the mid­Brunhes event (MBE) at ∼430 ka (Lüthi et al., 2008). The MBE was associated with the increase of atmospheric CO₂ and methane, as well as sea surface temperature (Barth et al., 2018). This event additionally coincided with the largest-amplitude change in δ¹⁸O in the global ocean over the past 6 Ma (Wang et al., 2003). Many authors have reported an abrupt increase in temperature after 424 ka (Oppo et al., 1998; de Abreu et al., 2005; Stein et al., 2009; Voelker et al., 2010; Rodrigues et al., 2011; Kandiano et al., 2012). We also see a peak in the TetraEther indeX (TEX₈₆) based the sea surface temperature (SST) record during MIS 11 in the western Bay of Bengal (Clemens et al., 2021; Fig. 4). This sudden shift is very prominent compared to the glacial-interglacial temperature changes prior to MIS 11. Globigerinoides ruber (pink) δ¹⁸O is the lowest among all the mixed layer species (Chiessi et al., 2007; Steph et al., 2009). The offset in δ¹⁸O between the two chromotypes of G. ruber could be attributed to the calcification of G. ruber (pink) at a warmer temperature than G. ruber (white) (Anand et al., 2003; Richey et al., 2012). Therefore, the significant increase in SST in the Bay of Bengal during MIS 11 made it suitable for the survival of G. ruber (pink). However, G. ruber (pink) must have been introduced in the northern Indian Ocean from a basin wherein it was thriving.

At present, the occurrence of G. ruber (pink) is limited to only the central Atlantic Ocean and nearby seas (Aurahs et al., 2009; Fig. 1). Its abundance decreases towards the north and south because of its warm water preference. That might be a reason why it is unable to reinvade the Indian and the Pacific Ocean through the southern tips of Africa and South America (Aurahs et al., 2009; Morard et al., 2019). The South Atlantic Ocean receives warm surface water from the Indian Ocean through the Agulhas current by the shedding of warm water “rings” and the advection of warm water filaments (Gordon et al., 1992; Lutjeharms, 1996, 2006; Fig. 1). Also, the slowdown of Atlantic Meridional Overturning Circulation warms the southeast Atlantic as the northward advection of subtropical Indian Ocean water masses accumulating in the South Atlantic Ocean is weakened (Wefer et al., 1996; Rühlemann et al., 1999, 2004; Dickson et al., 2010). Dickson et al. (2010) reported a sudden increase in SST in the southeast Atlantic Ocean and related it to the rapid release of Agulhas leakage waters and the increased rate of Agulhas ring shedding during MIS 11.

Thompson et al. (1979) related the last occurrence of G. ruber (pink) at MIS 5e to the genetic difference between the Atlantic and Indo-Pacific species populations. They further stated that the genetic difference might be due to several factors: the migration barrier of the Agulhas Current at the southern tip of Africa, high-latitude location of the southern boundary of Africa, the closing of the Isthmus of Panama, and the extension of South America to 55°S. These neo-tectonic activities occurred prior to the G. ruber (pink) FO we found in the Bay of Bengal, thus ruling out the possibility of any of these being the causal factor for FO of G. ruber (pink) in the northern Indian Ocean. In the South China Sea, Bühring et al. (2004) also reported G. ruber (pink) FO in MIS 11, in line with that reported from the northern Indian Ocean.

The other probable way by which G. ruber (pink) could invade the Indian Ocean is through the Indonesian Throughflow (ITF). The ITF brings the warm western Pacific Ocean water to the Indian Ocean, thereby modulating the oceanography of the northern Indian Ocean to a large extent (Xie et al., 2019). There have been reports of increased ITF influx into the southeastern Indian Ocean during the interglacials after the MPT, which increased the Agulhas leakage as well (Petrick et al., 2019). However, the possibility of G. ruber (pink) introduction in the Indian Ocean through the ITF is remote due to several reasons. Globigerinoides ruber (pink) appeared and disappeared from the Pacific Ocean at a similar time as that from the Indian Ocean (Thompson et al., 1979). Therefore, G. ruber (pink) must have been introduced into the Pacific Ocean prior to its subsequent invasion into the Indian Ocean by the ITF. The first occurrence of G. ruber (pink) has been reported comparatively earlier in the western Atlantic Ocean (1.04 Ma; Nishi et al., 2000) than in the eastern Atlantic Ocean (∼800 ka; Pflaumann, 1986), implying the possibility of an early invasion in the Pacific Ocean. However, the western Atlantic to the Pacific Ocean gateway is in a relatively colder, more southerly latitude than that between the eastern Atlantic and the Indian Ocean. A more southerly gateway implies a rare G. ruber (pink) population and thus reduced possibility of its invasion in the Pacific Ocean. The invasion through the Agulhas is further supported by the earlier occurrence in the Arabian Sea as compared to that in the Bay of Bengal. Considering these, we propose that G. ruber (pink) populated the Indian Ocean through the southern tip of Africa and not through the Indonesian Throughflow.

Thompson et al. (1979) related the FO of G. ruber (pink) in the middle Pleistocene to the progressive improvement in the preservation of the shell pigment and not to any abrupt evolutionary appearance. It has been reported that the pink colour fades with age, and it is difficult to differentiate the types of G. ruber in sediments older than 750 ka. However, the FO of G. ruber (pink) in our study is younger than that. Nevertheless, there is visible fading of the pink-coloured stain in the shells of G. ruber (pink) from the Arabian Sea prior to ∼426 ka. It is only at ∼426 ka when the pink colouration becomes prominent. However, this is not observed in the Bay of Bengal. The acme zone of G. ruber (pink) is post MIS 11 in both the Arabian Sea and Bay of Bengal as well as in the South China Sea. All these basins have entirely different hydrography and thus preservation potential of biogenic carbonates. A nearly similar acme zone in all three basins, despite the contrasting preservation potential, clearly rules out the possibility that fading of pink colouration with time due to poor preservation is the reason for the FO of G. ruber (pink) during MIS 11 in the northern Indian Ocean.

The first and last occurrence of G. ruber (pink) might also be closely related with the ecology of the symbiont responsible for its colouration. Pelagodinium béii is the symbiotic dinoflagellate present in the shells of G. ruber (white and pink), T. sacculifer and G. conglobatus (Gast & Caron, 1996; Takagi et al., 2019). The morphologic and metabolic relations between G. ruber (pink) and the dinoflagellate it harbours is still unclear. However, the laboratory experiments indicate that the symbionts aid in the calcification of T. sacculifer, and gametogenesis occurs when the symbionts are eliminated (Bé et al., 1982). Trilobatus sacculifer migrates to deeper cold waters during gametogenesis (Bé, 1980; Duplessy et al., 1981), indicating that the symbionts are sensitive to changes in temperature. However, only the symbiotic dinoflagellate harboured by G. ruber (pink) contains the pigment pheophytin, which gives it a pink colouration. Again, this could be because of the different ecological preferences of G. ruber (pink), compared to G. ruber (white) and T. sacculifer.

Similar symbionts are also reported from corals. Both planktic foraminifera and corals have dinoflagellate symbionts. Pelagodinium, a symbiont dinoflagellate commonly associated with planktic foraminifera, is closely related to the Symbiodinium, another dinoflagellate genus often inhabiting corals, as well as foraminifera (Decelle et al., 2015). The ecology of coral symbionts can provide clues to the ecology of the dinoflagellate associated with G. ruber (pink), and thus its FO and LO. Coral bleaching occurs due to the death of the symbiotic pigmented zooxanthellae (dinoflagellate), primarily because of the rise in temperature (Glynn, 1996; Brown, 1997). Unlike the coral that harbour zooxanthellae, non-zooxanthellate corals are not narrowly restricted ecologically (Stanley & Cairns, 1988). Therefore, the restricted ecology of symbiotic corals is because of the high sensitivity of the symbiotic dinoflagellate. Apart from temperature change, other major factors controlling coral bleaching are light availability and low salinity (Brown, 1997; Coles & Jokiel, 2018). The sensitivity of the symbiotic zooxanthellae in G. ruber (pink) could be related to that of the corals. Thus, a change in the ambient parameters would affect it greatly.

After MIS 11, SST decreased and δ¹⁸O_sw increased gradually until MIS 5e (Clemens et al., 2021). Globigerinoides ruber (pink) inhabited the Bay of Bengal and Arabian Sea until MIS 5e (128 ka), disappearing before the optimum habitable condition could be reached again. The low δ¹⁸O_sw value indicates low salinity, which could have resulted from high riverine influx caused by the strengthened monsoon. The resulting turbidity due to the high terrestrial influx again might have a negative effect on the symbionts of G. ruber (pink). With the same argument, G. ruber (white) and G. conglobatus, having similar dinoflagellate symbionts as G. ruber (pink) (Takagi et al., 2019), should also face the same consequences as G. ruber (pink) over the late Pleistocene in the Indian Ocean. The difference in the response of various symbiont-bearing planktic foraminifera is attributed to the variation in their ecological preferences.

Considering all of the above factors, we suggest that the warming of the southwest Atlantic Ocean enabled the pink genotype of G. ruber to invade the western Indian Ocean. It then made its way to the Bay of Bengal through the EICC during MIS 11 when the ambient conditions were suitable. The difference in hydrographic conditions between the Arabian Sea and the Bay of Bengal resulted in the first occurrence of G. ruber (pink) at different times in these two basins. The abrupt change in the surface hydrography during MIS 11 enabled the first occurrence of the pink genotype in the Bay of Bengal.

The LO and FO of G. ruber (pink) in the Bay of Bengal are 128 ka and 399 ka, respectively. In the Arabian Sea, the LO of G. ruber is at 123 ka. We propose that the warming of the southeast Atlantic Ocean enabled G. ruber (pink) to invade the Indian Ocean, and through the EICC, it reached the Bay of Bengal at 399 ka. The abrupt change in the hydrodynamic conditions in the Bay of Bengal helped this process by providing suitable conditions for G. ruber (pink) to inhabit. We rule out fading of the pink pigment present in G. ruber (pink) as a reason for its absence prior to 399 ka. Globigerinoides ruber (pink) disappeared from the Bay of Bengal at 128 ka because of the low temperature, low salinity, and high turbidity. We suggest that the FO of G. ruber (pink) can also be used as a biostratigraphic marker in the northern Indian Ocean.

The authors are thankful to the Director, National Institute of Oceanography, Goa, for his support and permission to publish this work. The authors are also thankful to the Council of Scientific and Industrial Research, the Government of India, and the National Centre for Polar and Ocean Research, Goa, for financial support. The International Ocean Discovery Program is acknowledged for drilling and providing the samples. Appendices 1 and 2 can be found linked to the online version of this article.

Details of sample interval, depth, age, number and relative abundance of white and pink varieties of planktic foraminifera Globigerinoides ruber at Site U1457 in the northeastern Arabian Sea.

Details of sample interval, depth, age, number and relative abundance of white and pink varieties of planktic foraminifera Globigerinoides ruber at Site U1446 in the northwestern Bay of Bengal.