Earth Science India Vol.1 (IV), October, 2008, pp.243-257
e-journal: http://www.earthscienceindia.info
Marine 14C reservoir age and Suess effect in the Indian Ocean
 
Koushik Dutta
AMS Radiocarbon Laboratory, Institute of Physics
Sachivalaya Marg, P.O. Sainik School, Bhubaneswar – 751 005, India

Abstract: Apparent radiocarbon (14C) ages of marine biogenic samples that derive their carbon from surface seawater dissolved inorganic carbon are on the average about 400 years older than contemporary terrestrial woods or global atmospheric CO2. This age offset is due to mixing of old carbon from the deep ocean and is referred to marine 14C reservoir effect. Both regional and temporal variations of ocean circulation pattern causes significant spatial and temporal variations in marine 14C reservoir ages and hence of biogenic surface marine samples. Knowledge of reservoir age is very important to accurately calibrate 14C-ages of biogenic carbonates and sediment organic matter from marine sediments that are frequently used in paleoceanographic studies. Here the concept of marine 14C reservoir ages and their quantification are discussed and data for the Indian Ocean region are reviewed. Regional variations of marine 14C reservoir ages and fossil fuel Suess effect for the Indian Ocean and the South China Sea are analyzed.

Introduction to 14C in the environment

           The only naturally occurring radioactive isotope of carbon is 14C, commonly known as radiocarbon. 14C is introduced in the environment through its production in the upper atmosphere by the reaction of atmospheric nitrogen with thermal neutrons that are produced form cosmic ray spallation reaction on other atmospheric components.
14N7 + 1n0 = 14C6 + 1H1 + (energy)
 
         The energetic 14C atoms freshly produced in the atmosphere are soon oxidized in presence of atmospheric oxygen to form carbon monoxide which eventually oxidises to carbon dioxide.
  14C6 + O = 14CO
 14CO + O = 14CO2

           Radioactive carbon dioxide 14CO2 being indistinguishable from other forms of CO2 (12CO2 or 13CO2), it eventually enters the biosphere through photosynthesis by terrestrial plants, and then to the entire food chain through herbivorous animals. 14C activities of most terrestrial living organisms are therefore in equilibrium with that in the atmosphere, through continuous exchange of 14C by photosynthesis or food intake and respiration. When an organism dies the 14C exchange halts, and the 14C in the dead tissues start to decrease exponentially through radioactive decay. 14C forms stable nitrogen through beta decay and half-life of 5730±40 years.
 
14C6 = 14N7 + β– 
(T½ = 5730±40 years)

         There are various conventions for expressing 14C concentrations in natural samples (Stuiver and Polach, 1977; Mook and van der Plicht, 1999). 14C concentration in a sample can be expressed either as age in years before present or BP, where 'present' is actually the year AD 1950. For geochemical applications 14C is commonly expressed as Δ14C which is permil (‰) deviation of 14C/12C ratio of the sample with respect to a modern 14C standard (whose 14C activity equal to that of pre-industrial wood of AD 1890), the ratios being normalised for isotopic fractionation of the 13C isotope and corrected for decay of 14C. Old carbon containing samples (such as foraminifers from deep sea cores) have less 14C than that present in a modern sample and therefore have positive 14C-ages in BP and negative Δ14C ‰ values; whereas most present day terrestrial wood samples with higher 14C than the modern 14C standard show negative 14C-ages in BP and positive Δ14C ‰ values. Geologically old carbon samples (such as marble or old limestone, coal and other fossil fuels) are '14C-dead' from which nearly all 14C got decayed away. Such samples have infinite 14C-age and Δ14C close to –1000‰.

          Oceans are important reservoir of exchangeable carbon (40,000 Gt), most of which are present as dissolved inorganic carbon (DIC). The natural steady-state concentration of 14C in the DIC of surface ocean waters is in quasi-equilibrium with atmospheric 14C production, air-sea exchange of atmospheric CO2, and mixing with deeper ocean water (Fig-1). The mixing time-scale of the global oceans through large-scale thermohaline circulation is about 1000 years. This causes aging of a given water mass during its residence in the ocean since its last equilibration with the atmosphere when it was closer to the surface, and results in 14C-ages of deep ocean water close to 1000 BP (Δ14C about –200‰) (Bien et al., 1965; Stuiver and Quay, 1983). Diffusive and advective mixing with 14C-depleted deep ocean waters causes the 14C concentration for the surface oceans to be 5% lower on the average than in the contemporary atmosphere. Hence the 14C-ages of surface ocean waters are apparently older than contemporary terrestrial plants by about 400 years (Taylor and Berger, 1967; Bard, 1988; Hughen et al., 2004). This phenomenon is known as marine 14C reservoir effect. While the long-term temporal variations of surface ocean 14C-ages are a function of the contemporary atmospheric 14C levels and vertical mixing rates of the global ocean, their spatial variations also depend on the regional circulation patterns within the thermocline and regional air-sea CO2 exchange rates. Local variations of marine 14C-ages can take place due to input of hard water whose DIC is depleted in 14C. Sources of hard water include rivers draining a geologically old carbonate terrain or ground water flowing through old carbonate aquifer. All these factors cause considerable spatial variations in the steady-sate (or natural) distribution of 14C in the surface oceans and regional offsets of marine reservoir 14C-ages from the global mean value, and make it difficult to use a common 14C calibration curve for 14C-dating of all marine samples. This is unlike 14C-dating of terrestrial samples, which depends on past 14C levels of rapidly mixing global atmosphere with a small difference between the two hemispheres.
 

Fig-1: Natural radiocarbon age distribution in the marine environment.

             The introduction of anthropogenic 14C produced from atmospheric nuclear tests of the late 1950’s and early 1960’s had nearly doubled the 14C levels of the atmosphere CO2 , which eventually transferred to other exchangeable carbon reservoirs as the biosphere and the oceans (Rafter and Fergusson, 1957). Mixing of bomb-14C contaminated atmospheric CO2 resulted in steady increase of the level of 14C in the surface oceans. This injection of bomb 14C in surface waters though obliterated the natural (pre-nuclear) 14C signatures and its inventory in the water column provided a means of determining the air-sea exchange rates of CO2. The results of state-of-the-art ocean general circulation models (GCMs), which predict the future rise of greenhouse gases in the atmosphere, must be verified through observations before they can be used for planning of policies to control anthropogenic CO2 emissions. Such models can be also used to simulate steady state distributions of pre-nuclear Δ14C for the global oceans (Maier-Reimer, 1993; Toggweiler et al., 1989). Comparison of these model results with direct measurements of 14C in surface dwelling marine calcareous organisms from the pre-nuclear era offer a way to check the validity of these models and testing their predictive capabilities.

            Knowledge about the natural (or pre-nuclear) level of 14C in the surface ocean waters is necessary to determine:

(i) Offsets of regional reservoir ages from the modelled global mean—to calibrate the 14C-ages of marine calcareous fossils, which grew during the pre-nuclear period and used for dating of marine sediments;

(ii) Temporal variations of bomb-14C in water column and its inventory—to assess the air-sea CO2 exchange rates and decadal circulation in the thermocline
             First oceanic measurements of 14C started during the mid 1950’s—about the same time when bomb-14C started penetrating the world oceans. Soon, it was realized that bomb-14C can be used as a powerful tracer to study air-sea exchange rates of CO2 and mixing in the thermocline region on decadal time scales, utilizing the temporal variations of 14C concentrations (Rafter and O’Brien, 1973; Broecker et al., 1978; Broecker and Peng, 1982; Broecker et al., 1995). The need for information about the pre-nuclear surface ocean 14C levels was felt, while attempting to determine the increase of the surface 14C levels from the pre-nuclear values. To determine the penetration depth and inventory of bomb-14C in oceans, it is essential to have knowledge about the surface pre-nuclear 14C (Broecker et al., 1985). Several approaches have been made to determine these pre-nuclear surface ocean 14C values from the analyses of suitable marine samples from the pre-nuclear era. Broecker et al., (1985) estimated the pre-nuclear Δ14C values for different oceanic regions of the world within ±10‰. These values range from a minimum of –140‰ in the polar oceans to –50‰ in the mid latitudes. However, the surface ocean circulations for the global oceans are far from simple to adopt a uniform scheme for meridional pre-nuclear Δ14C variations for different oceans.

Chronological applications of 14C

             Since carbon is ubiquitous in all living beings, the radioactive decay of 14C can be conveniently used for age determination of organic or biogenic samples. The science of 14C-dating was established by Prof. Willard Frank Libby of the University of Chicago, who for the first time demonstrated the potential of 14C in dating of archaeological artefacts (Libby et al., 1949; Libby, 1955), and was awarded the Nobel Prize in Chemistry in 1960 for this pioneering discovery. The half-life of 14C (5730±40 years) is suitable for dating of samples that are within 50,000 years old, thus making it the most widely used dating method in Quaternary geochronology. To determine accumulation rates of marine sediments, the most common method in use is 14C-dating of planktonic foraminifers. Adult species (250 to 500 μ) of surface dwelling foraminifera, such as Globigerinoides ruber, Globigerinoides sacculifer, Neogloboquadrina dutertrie and Orbulina universa are best suited for 14C-dating, which grow within the top 100 m of the seawater column. In special cases such as in near shore areas where planktonic species are scarce, calcareous benthic foraminifera species may also be used. Marine organic matter may also be used for 14C-dating of sediments when calcareous fossils are totally absent. However, as will be discussed later the last two methods may need special attention for their 14C-age calibration. All samples chosen for analysis are carefully scrutinized and are thoroughly cleaned from all forms of contamination which may potentially alter their 14C-age. Since the amount of datable carbon from sediments are usually small—typically few tens of milligrams from a gram of sediment, accelerator mass spectrometry (AMS) method is essential for 14C analysis of these samples. For accurate calibration of marine 14C-dates their initial 14C-ages must be well known, that depend on regional marine reservoir effect.

Concept of marine 14C reservoir age and ΔR correction values

               For a given oceanic region at any given time, the 14C reservoir age is defined as the difference between the measured conventional 14C-age of the reservoir (seawater) and that of the contemporary atmospheric CO2 (derived from tree-rings). 14C reservoir ages can be determined from 14C analysis of pre-nuclear (pre-1950s) marine calcareous samples of known calendar age. Considerable spatial variability is seen in the regional marine reservoir 14C-ages from the global mean value, mainly due to variations in local ocean circulation patterns and regional exchange rates of CO2 with the atmosphere. For a given region at any given time, the difference between the regional marine 14C-age and that of the modelled global surface ocean is expressed by the term ΔR (pronounced as 'delta-R'), which account for regional deviations of reservoir ages from the global mean value (Stuiver and Braziunas, 1993). The regional ΔR correction values can be determined from the difference of regional 14C reservoir age and that of the modelled global ocean.

              Marine reservoir 14C-ages exhibit considerable spatial variations due to change in ocean circulation regimes. In those oceanic regions where favourable conditions exist for mixing with deeper ocean water (with older 14C-ages) through wind induced upwelling, surface marine reservoir 14C-ages will be higher than the global average value. Therefore ΔR for such regions will have a positive value. Examples of such region in the Indian Ocean are the eastern and the western Arabian Sea, where upwelling takes place through seasonal monsoon induced winds. DIC inputs from rivers that leach geologically old carbonate rocks may result in large regional reservoir 14C-ages (Little, 1993). On the other hand, 14C in seawater DIC are higher than the average global ocean in shallow oceanic regions or in enclosed lagoons, where mixing with 14C depleted water from deeper ocean is limited. Such regions have negative ΔR values, examples of which include the Gulf of Kutch in the northeastern Arabian Sea, and the Chilika Lake in the northern Bay of Bengal. Shallow oceanic regions of the western Pacific such as the southern part of the South China Sea also have negative ΔR values, where the water depth is less than 600 m. According to the latest compilation of Global Marine Reservoir Correction Database (Reimer and Reimer, 2001), ΔR values for the global oceanic regions excluding the Southern Ocean vary from –200 year to +800 year, and majority of them fall between –50 years to +250 years. As discussed later, rapid vertical mixing off the Antarctica coast causes unusually high ΔR values for the Southern Ocean.

Fig-2: Illustration of the concept of marine reservoir effect and ΔR correction values. (a) The Marine04 calibration curve (Hughen et al., 2004). (b) Reconstructed marine 14C-ages at three hypothetical oceanic regions. Top curve: in an upwelling region for which reservoir 14C-age is older than the global mean (ΔR = +100 years). Middle curve: in an oceanic area that resembles the average world ocean (ΔR = 0 year). Bottom curve: an oceanic area with lower vertical mixing with deeper waters where reservoir 14C-age is younger than the global mean (ΔR = –50 years).

            The Marine04 calibration curve (Hughen et al., 2004), is shown in Fig-2(a) which has been determined using the decadal tree-ring 14C data and the carbon cycle model after Oescheger et al., (1975). Regional ΔR correction values must be added to this curve in Fig-2(a) to obtain the regional marine 14C calibration curves shown in Fig-2(b). Reconstruction of regional marine 14C-age calibration curves for three hypothetical oceanic regions are shown in Fig-2(b). The Marine04 calibration curve extends back to 26,000 years BP. For calibrating older 14C-dates up to 50,000 years BP one may use the calibration curve after Fairbanks et al., (2005), which uses high-precision 14C data of very old unaltered corals absolute dated with U-Th series isotopic method.

              Knowledge of regional ΔR values is necessary for accurate calibration of 14C-ages of marine samples. It is implicit in the definition of ΔR, that temporal variations of regional reservoir ages will parallel those of the global ocean, thus ΔR is assumed to be time independent for any given region (Stuiver et al., 1986, 1998). The reservoir 14C-ages and ΔR correction values reported from various oceanic regions are valid for the modern state of ocean circulation, since these were usually derived from samples that grown within past few hundred years or so. Since climate induced changes in ocean circulation may effect both mixing with deep water and atmosphere-ocean CO2 exchange, regional reservoir 14C-ages may respond to major shifts in climate, thus temporal variations of ΔR is inevitable. Reservoir 14C-ages change at glacial-interglacial time scales, due to major shifts in ocean circulation pattern. Model simulations of past reservoir 14C-ages have indicated that 30% reduction in Atlantic meridional circulation may enhance reservoir ages in high latitudes by 500 years (Franke et al., 2008). If the changes in regional reservoir 14C-ages are significantly different from that of the global modelled ocean, the ΔR values would also change. Thus the assumption that ΔR values are constant for a given region is not strictly valid. Staubwassar et al., (2002) and Ascough et al., (2004, 2006) reported such changes indeed took place during the Holocene in the Arabian Sea and in the north Atlantic Ocean.

Determination of marine 14C reservoir effect and regional ΔR values

              An ideal approach to determine the 14C-age of seawater is to measure directly 14C in its DIC (Fonselius and Östlund, 1959). However, direct measurements of 14C in oceans began too late—only after the onset of nuclear weapons testing in the early 1950’s (Rubin and Suess, 1955). Therefore nearly all the seawater samples collected for 14C determinations were contaminated with bomb 14C. To determine pre-nuclear surface ocean Δ14C, reservoir 14C-ages and ΔR correction values, one must rely on the analysis 14C in marine samples of known age, which were grown or collected before 1950. Some of the commonly used methods for determination of marine 14C reservoir ages and ΔR correction values are described here.

(a) Marine shells: Measurement of 14C in archived marine calcareous shells (e.g., bivalves or gastropods) collected prior to the nuclear-testing era is the most common approach to determine marine 14C reservoir ages or pre-nuclear levels of surface water Δ14C (Stuiver et al., 1986; Siani et al., 2000; Dutta et al., 2001). Short-lived and epifaunal (non-burrowing) mollusc species are preferred with well-documented collection dates. One can also measure 14C in long-lived giant calms (e.g., Tridacna gigas or Arctica islandica), which form well-developed annual growth bands (Weidman and Jones, 1993). Shell derived marine 14C reservoir ages often yield anomalous old ages due to incorporation of geologically old detrital carbonates in the shells (Dye, 1994). Several studies have indicated that 14C ages of various mollusc species differ due to difference in their habitat and diet. However, Ascough et al., (2005a) concluded that any mollusc species may be suitable for the purpose if no geologically old carbonate rocks or sources of hard water are present.

(b) Corals: Hermatypic or reef building corals form annual growth bands made of aragonite (CaCO3) with 14C/12C in equilibrium with the ambient seawater DIC. Once deposited these aragonite bands do not exchange their carbonate with external source unless they are very old causing recrystallisation. In a live collected coral these bands can be precisely dated back in time by counting them just as in tree-rings. Past changes in 14C/12C ratio of the DIC of surface-oceans can be determined from 14C measurements in coral growth bands. Some of the earliest 14C measurements in coral skeletons were reported by Moore et al., (1973), Moore and Krishnaswami (1974) and Buddemeier et al., (1974). Continuous records of 14C variations in surface oceans can be deciphered from long corals, extending back to several hundred years in the pre-nuclear era (Nozaki et al., 1978; Druffel and Linick, 1978; Druffel, 1981; Druffel and Suess, 1983; Druffel and Griffin, 1993; Grumet et al., 2002; Hua et al., 2004). The major disadvantage of using corals is their restricted geographical distribution. Their habitat is confined mainly within the tropical or limited sub-tropical oceanic regions.

(c) Charcoal and marine shell pairs: Direct determination of the atmosphere-ocean 14C-age difference can be obtained from 14C measurements in co-existing terrestrial charcoal and marine shell pairs collected from same stratigraphic horizon, thus capturing the oceanic and terrestrial events simultaneously (Southon et al., 1990, 1992; Talma, 1990; Little, 1993; Ingram, 1998). This method has potential to determine past variations of regional reservoir ages in response to the ocean circulation changes for longer timescales. To measure past reservoir ages, Bard (1988) had suggested comparing the 14C-age of a short and well-defined marker or event both on the land and in the deep-sea sediments (e.g., deposition of a volcanic ash layer).

(d) Otoliths: Otoliths are hard calcareous deposits that grow within the inner ears of teleost or bony fish, which primarily function as gravity and auditory receptors. Similar to corals, otoliths are made of aragonite and form annual growth layers. AMS 14C measurements in otoliths of known age marine fishes can be used for reconstructing the 14C evolution in surface water (Higham and Hogg, 1995; Kalish, 1993; Kalish et al., 2000). Advantage of using fish otoliths over other methods is their large geographical range. Fish being nektonic (or free swimming), their otoliths can integrate the past 14C signals over much wider oceanic region, unlike other archives as in shells or corals that record oceanic 14C variations only at their growth location.

           Most of the methods described above have their own merits and demerits, and are often fraught with problems. Ascough et al. (2005b) reviewed on various intricacies involved in determination of marine reservoir 14C-ages. Reservoir ages and ΔR correction values are usually applicable for surface marine samples, but may not be valid for samples such as benthic foraminifers which live at the sediment surface. However, the same ΔR correction values may be applicable for benthic samples too, if the water depth of their occurrence is within 50 m or so as often the case for near shore samples. Reservoir ages derived from surface marine biogenic carbonates are also applicable to marine organic matter, since they derive their carbon from the same reservoir (seawater DIC). However, complications may arise if the organic matter is a mixture of terrestrial and marine organic matters. Organic molecules of purely marine origin will register the marine reservoir effects in their 14C-ages, while terrestrial organic matters may show a range of ages that can be either too old or too young. In such cases, it may be beneficial to assess the contribution of each organic matter types from geochemical (C/N wt.) or isotopic (δ13C) ratios, or even better to isolate individual organic compounds of purely marine origin. All reservoir 14C-age estimates are long-term average, ignoring any seasonal DIC 14C changes that may occur due to seasonal change in ocean circulation pattern. Most known age marine biogenic carbonate samples used for regional reservoir age determinations are usually collected from coastal locations or small islands, which are then extrapolated for use with open ocean samples. Reservoir age corrections as derived using samples from small islands found to match fairly well to that derived from nearby coastal locations, since the surface waters of the coastal ocean are normally get replenished by the open ocean water in a short time scale.

          In spite of the availability of several established methods, pre-nuclear surface water 14C data for many key oceanic regions are still sparse mainly due to scarcity of suitable known-age marine samples. The Indian Ocean region is an area of considerable interest from the perspective of paleoclimate and paleoceanographic studies. Unfortunately spatial data of 14C reservoir ages for this region is too meagre, essential for reporting reliable 14C chronologies crucial for such studies.
 
Studies of reservoir 14C-ages and ΔR correction values for the Indian Ocean

                  The earliest seawater DIC 14C measurements in the tropical Indian Ocean region were done during the expeditions named Monsoon (10°S, 99°E; November, 1960), Lusiad (8°N, 71°E; October, 1962) and Dodo (6°S, 55°E; September, 1964) by the Scripps Institute of Oceanography, USA (Linick, 1975). The surface ocean Δ14C values measured during these periods ranged from –38 to –24‰, for the latitude band of 24°S to 11°S during the Monsoon expedition. These results, however, could not provide pre-nuclear Δ14C values for the region. These values are significantly higher than those obtained from pre-nuclear Δ14C in biological archives—indicating the penetration of bomb 14C even during early 1960s. More extensive studies of surface water Δ14C measurements were made between 1977 and 1978 as a part of the international GEOSECS Indian Ocean expedition (Stuiver and Östlund, 1983). To calculate the bomb 14C inventories from the GEOSECS 14C measurements, Broecker et al., (1985) assumed pre-nuclear surface Δ14C values of –60 to –65‰ for the northern Indian Ocean. In a later report, Broecker et al., (1995) assumed surface natural Δ14C –59‰ in the equatorial Indian Ocean (GEOSECS 448), and a steadily decreasing Δ14C trend was followed towards the north, assuming minimum Δ14C value of –68‰ for the northern Arabian Sea (GEOSECS 416). Moore and Krishnaswami (1974) first reported measurement of 14C in the growth bands of corals from the Gulf of Kutch. Δ14C of coral growth bands grown during or before 1950s have been reported for corals from the Red Sea and Gulf of Aden (Cember, 1989), the Gulf of Kutch (Chakraborty, 1993) and the western Indian Ocean (Grumet et al., 2002).

                First systematic studies of marine 14C-reservoir ages and ΔR correction values for the northern Indian Ocean were reported by Dutta et al., (2001), from 14C measurements in archived marine mollusc shells collected between 1930 and 1954. The ΔR correction values were calculated from the 14C data of annual growth bands of corals from the Gulf of Kutch (Chakraborty, 1993; Bhushan et al., 1994), and AMS 14C dates of annually laminated sediments in the northeastern Arabian Sea (von Rad et al., 1999). Southon et al., (2002) reported ΔR correction values for several locations in the Indian Ocean and the South China Sea mainly using museum archived marine shells. Hua et al., (2004) reported ΔR correction values for the Cocos (Keeling) Islands in the eastern Indian Ocean using annual coral bands.

Spatial and temporal variations of marine 14C reservoir ages in the Indian Ocean

The data of marine 14C reservoir ages and ΔR correction values for the Indian Ocean and South China Sea are compiled here and shown in Fig-3. The 14C-reservoir ages for the open eastern Arabian Sea range from 622 to 390 years, which are in general higher when compared to the open Bay of Bengal values which range from 315 to 474 years. Reservoir ages in the shallow marine areas of the Gulf of Kutch and the Chilika Lake lagoon are 241 years and 229 years, respectively (Dutta et al., 2001). One sample collected near Pondicherry in the western Bay of Bengal yield anomalous high reservoir age of 831 years, possibly due to incorporation of geological carbonate (Southon et al., 2002). These two shallow marine samples and the anomalous old sample from Pondicherry were excluded for determining regional ΔR correction values of the Indian Ocean. The reservoir ages are younger in the shallow marine areas of the South China Sea. The Red Sea, the Persian Gulf, the western Arabian Sea and the western Indian Ocean near Madagascar have rather narrow range of reservoir 14C-ages, mostly lying between 400 and 500 years. The eastern Indian Ocean south of Indonesia and west of Australia has reservoir 14C-ages between 323 and 377 years, similar to the Bay of Bengal values.

 
Fig-3: Spatial variation of marine 14C reservoir ages for the entire Indian Ocean obtained from Global Marine Reservoir Correction Database (Reimer and Reimer, 2001). Reservoir 14C-ages (in years) reported from selected locations are shown, while regional average ΔR values (in years) are in given bold with 1-σ standard deviation. Indian Ocean reservoir 14C-age data compiled from Bowman (1985a, 1985b), Cember (1989), Bhushan et al., (1994), von Rad et al., (1999), Dutta et al., (2001), Southon et al., (2002), Grumet et al., (2002), and Hua et al., (2004). Values for the southern Indian Ocean south of 60°S (in parentheses) have been estimated from Berkman and Forman (1996).

               The observed pattern of 14C reservoir ages in the Indian Ocean can be explained in terms of regional variation of circulation patterns within the thermocline. In the northeastern Arabian Sea vertical mixing is favoured by (i) seasonal upwelling during southwest monsoon and (ii) convective processes associated with winter cooling (Madhupratap et al., 1996). The average reservoir 14C-ages for the Arabian Sea is older than that of the modelled world ocean (mean ΔR ~ 180 years), due to upwelling induced mixing with deeper 14C-depleted water derived from Antarctic Bottom Water from the south. The Bay of Bengal, which receives large amount of fresh water from several major rivers (Milliman and Mead, 1983) has steep gradients of the isopycnal (equal density) surfaces within the top 200 m. Steep isopycnal gradient in the Bay of Bengal may greatly impede the vertical mixing rate, retarding the advection of deeper 14C depleted water. This resulted in relatively younger reservoir 14C-ages (ΔR ~ 60 years) for the Bay of Bengal compared to the Arabian Sea. Input of riverine DIC depleted in 14C may tend to counteract the effect of reduced mixing (Little, 1993), due to the hard water effect. Higher reservoir ages in the Arabian Sea compared to the Bay of Bengal shows that greater vertical mixing is the dominant process, which determines the observed pattern of 14C reservoir ages in the northern Indian Ocean. Interestingly, the modelled pre-nuclear surface ocean 14C distribution by Toggweiler et al., (1989) and Guilderson et al., (2000) predicts younger 14C-ages in the Arabian Sea than in the Bay of Bengal. This is in contrary to the pattern observed from the 14C data of pre-nuclear marine carbonates.
              These ΔR values would be useful in calibrating the 14C-ages of marine microfossils for dating marine sediments and also for dating coastal calcareous archeological samples from the Indian Ocean region. The regional average ΔR correction values given in Fig-3 have taken into account all valid reservoir 14C-age data, and are only indicative of the large scale pattern. In order to calibrate a given marine 14C date, ΔR data from the nearest locations must be chosen. Pooled mean (error weighted) ΔR correction values from these locations must be used for the calibration.

             Significant climate induced changes in the northern Indian Ocean circulation took place during the glacial and interglacial periods (Duplessy, 1982). Since the marine 14C reservoir age of the northern Indian Ocean, particularly in the Arabian Sea is controlled by summer monsoon upwelling, winter monsoon convection and the reservoir age of the Arabian Sea thermocline water which is derived from Antarctic Bottom Water (Staubwasser et al., 2002), variations in any of these factors due to global climate shifts can potentially alter the regional reservoir 14C-ages. Staubwasser et al., (2002) reported that the surface Arabian Sea reservoir ages were more than 1000 years during the deglaciation, and they varied between 780 and 1120 years during the early Holocene.

Marine 14C reservoir ages of the southern Indian Ocean off Antarctica

             The southern Indian Ocean is a region for which knowledge of pre-nuclear surface water 14C is too meagre. The earliest seawater 14C measurements in the southern Indian Ocean region were compiled by Delibrius et al., (1980). Reports of marine reservoir 14C-ages close to Antarctica range from 932 BP to 1632 BP, corresponding to mean ΔR of 849±13 year (n=14). Till date, no systematic study on marine reservoir 14C-ages had been done specifically for the Southern Indian Ocean near Antarctica. This oceanic region is an area of considerable interest from the perspective of paleoclimate studies from multi-proxy geochemical analyses of marine sediments. Such studies must rely on firm 14C-chronology, for which accurate knowledge on reservoir 14C-age is essential. Berkman and Forman (1996) recommended use of reservoir 14C-ages of 1300±100 year for the Southern Ocean region. However, it is unlikely that reservoir age of the Southern Indian Ocean off Antarctica coast are significantly different than in other areas of the Southern Ocean, due to fast zonal mixing by the circum-polar currents.

Marine 14C Suess effect in the northern Indian Ocean and the South China Sea

              Following the beginning of industrial era during the late 19th century and before the onset of large scale atmospheric tests in the late 1950's, 14C in the atmospheric CO2 steadily decreased by about –25‰ through dilution by 14C depleted fossil fuel CO2. Invasion of fossil fuel CO2 to the oceans caused significant depletion of surface ocean Δ14C values between 1900 and early 1950s, an phenomena known as marine Suess effect (Druffel, 1981; Druffel and Suess, 1983; Toggweiler et al., 1989). The observed depletion was typically few permil per decade, but variable among different oceanic areas (Druffel and Griffin, 1993). The Δ14C values of the samples reported from the northern Indian Ocean and the South China Sea are analyzed in this study, which also show a decreasing trend between AD 1900 and 1955 (Fig-4). The results show the magnitude of marine 14C Suess effect of –3.0±1.8‰ per decade for the Arabian Sea, –2.1±0.5‰ per decade for the Bay of Bengal, and –2.0±1.3‰ per decade for the South China Sea.
 
Fig-4: Suess effect in the Northern Indian Ocean and the South China Sea. Data from Dutta et al., (2001) plotted as circles and from Southon et al., (2002) plotted as triangles and squares. Symbol fills: white – Bay of Bengal, black – Arabian Sea; and grey – South China Sea. Straight lines indicate linear fit to the Δ14C data between 1900 and 1955.

              Even though the uncertainties in individual reservoir age estimates are of the order of ±5‰, several samples were considered to evaluate the Suess effect for a large oceanic area. Thus, the small values of the Suess effect (2 to 3±1‰) reported in this study are statistically significant. Magnitude of Suess effect for various ocean basins depends mainly on the penetration of anthropogenic CO2 to the ocean. The present analysis shows similar magnitude of Suess effect for the Arabian Sea and the Bay of Bengal, as indicated by the parallel trends of the linear fits, though the vertical scales of these lines are offset depending on their reservoir ages.

               Further measurements from the Indian Ocean region are needed to assess and reduce the uncertainty in marine 14C Suess effect. This can be better understood from 14C analysis in long coral cores dating back to the pre-industrial era. Druffel (1981) reported post-industrial and pre-nuclear Δ14C recorded in corals from the Urvina Bay, Galapagos Island, in the eastern equatorial Pacific Ocean, which also show similar magnitude of Suess effect of about –6‰ between 1930 and 1953.

Conclusions

            14C is commonly used for dating of marine sediments that are up to 50,000 years old. Initial 14C-ages depend on regional marine 14C-ages of surface ocean and any variations thereof must be known to accurately determine 14C-dates of marine samples. Information on natural or pre-bomb 14C in surface oceans is also important for studies of oceanic carbon cycle that use anthropogenic or bomb-14C as a tracer. Reservoir 14C-ages has been reported from many places in the Indian Ocean analyzing known age marine biogenic carbonates, but data on its spatial and temporal variability is still not adequate. The Δ14C values between 1900 and 1955 from the Indian Ocean and the South China Sea show a decreasing trend, indicating marine 14C Suess effect in the Bay of Bengal. The average magnitude of this Suess effect is –3±2‰ per decade for the Arabian Sea and – 2±1‰ per decade for the Bay of Bengal and the South China Sea.

Internet resources on marine 14C reservoir ages and marine 14C-age calibration

              Interested readers may find the following internet resources useful, for further information related to 14C-dating and marine 14C-calibration. Pre-nuclear 14C measurements for the world oceans can be obtained from Global Marine Reservoir Correction Database, maintained jointly by the Queen’s University of Belfast, UK and the University of Washington, Seattle, USA (Reimer and Reimer, 2001).
Marine Reservoir Correction Database (Queen's University Belfast, UK): http://intcal.qub.ac.uk/marine
Radiocarbon Reservoir Ages Database (University of Bremen, Germany): http://reservoirage.palmod.uni-bremen.de
Richard G. Fairbanks Radiocarbon Calibration (LDEO, Columbia University, USA): http://radiocarbon.ldeo.columbia.edu

(Above web site links were accessed on Sep 30 2008)
 
ACKNOWLEDGEMENTS: A portion of this research work was carried out at Physical Research Laboratory, Ahmedabad, with a research fellowship from Dept. of Space, Govt. of India.

References
Ascough, P.L., Cook, G.T., Dugmore, A.J., Barber, J., Higney, E., and Scott, E.M. (2004) Holocene variations in the Scottish marine radiocarbon reservoir effect. Radiocarbon, v.46, pp.611-620.
Ascough, P.L., Cook, G.T., Dugmore, A.J., Scott, E.M. and Freeman, S.P.H.T. (2005a) Influence of mollusk species on marine ΔR determinations. Radiocarbon, v.46, pp.611-620.
Ascough, P., Cook, G., and Dugmore, A. (2005b) Methodological approaches to determining the marine radiocarbon reservoir effect. Prog. Phys. Geogr., v.29, 532, doi: 10.1191/0309133305pp461ra.
Ascough, P.L., Cook, G.T., Church, M.J., Dugmore, A.J., Arge, S.V., and McGovern, T.H. (2006) Variability in North Atlantic marine radiocarbon reservoir effects at c. AD 1000. The Holocene, v.16, 131, doi: 10.1191/0959683606hl913rr.
Bard, E. (1988) Correction of accelerator mass spectrometry 14C-ages measured in planktonic foraminifera: Paleoceanographic implications. Paleoceanogr., v.3, pp.635–645.
Berkman, P.A., and Forman, S.L. (1996) Pre-bomb radiocarbon and the reservoir correction for calcareous marine species in the southern ocean. Geophys. Res. Lett., v.23, pp.363-366.
Bien, G.S., Rakestraw, N.W., and Suess, H. (1965) Radiocarbon concentration in the Pacific and Indian Oceans and its relation to deep water movements. Limnology and Oceanogr., v.10, pp.25–37.
Bhushan, R., Chakraborty, S., and Krishnaswami, S. (1994) Physical Research Laboratory (Chemistry) radiocarbon date list I. Radiocarbon, v.36, pp.251–256.
Bowman, G.M. (1985a) Revised radiocarbon oceanic reservoir correction for Southern Australia. Search, v.16, pp.164–165.
Bowman, G.M. (1985b) Oceanic reservoir correction for marine radiocarbon dates from northwestern Australia. Aus. Arch., v.20, pp.58–67.
Broecker, W.S., Peng T.–H., and Stuiver, M. (1978) An estimate of the upwelling rate in the equatorial Atlantic based on the distribution of bomb radiocarbon. Jnl. Geophys. Res., v.83, pp.6,179–6,186.
Broecker, W.S., Peng, T.–H. (1982) Tracers in the Sea. Lamont Doherty Geological Observatory Press, Palisades, New York.
Broecker, W.S., Peng, T.–H., Östlund, H.G., and Stuiver, M. (1985) The distribution of bomb radiocarbon in the ocean. Jnl. Geophys. Res., v.90, pp.6,953–6,970.
Broecker, W.S., Sutherland, S. Smethie, W., Peng, T.–H., and Östlund, H.G. (1995) Oceanic radiocarbon: Separation of the natural and bomb components. Glob. Biogeochem. Cyc., v.9, pp.263–288.
Buddemeier, R.W., Maragos, J.E., and Knutson, D.W. (1974) Radiographic studies of reef coral exoskeletons: rates and patterns of coral growth. Jnl. Exp. Mar. Bio. and Ecol., v.14, pp.179–200.
Cember, R. (1989) Bomb radiocarbon in the Red Sea: A medium-scale gas exchange experiment. Jnl. Geophys. Res., v.94, pp.2,111–2,123.
Chakraborty, S. (1993) Environmental significance of isotopic and trace elemental variations in banded corals. Ph.D.–Thesis, MS University of Baroda, Vadodara, India.
Delibrias, G. (1980) Carbon-14 in the Southern Indian Ocean. Radiocarbon, v.22, pp.684–692.
Druffel, E.R.M., and Linick, T.W. (1978) Radiocarbon in annual coral rings of Florida. Geophys. Res. Lett., v.5, pp.913–916.
Druffel, E.R.M. (1981) Radiocarbon in annual coral rings from the eastern tropical Pacific Ocean. Geophys. Res. Lett., v.8, pp.59–62.
Druffel, E.R.M., and Suess, H. (1983) On the radiocarbon record in the banded corals: Exchange parameters and net transport of 14CO2 between atmosphere and surface ocean. Jnl. Geophys. Res., v.88, 1,271-1,280.
Druffel, E.R.M., and Griffin, S. (1993) Large variations of surface ocean radiocarbon: Evidence of circulation changes in the southwestern Pacific. Jnl. Geophys. Res., v.98, pp.20,249–20,259.
Dupplessy, J.–C. (1982) Glacial to interglacial contrasts in the northern Indian Ocean. Nature, v.295, pp.494–498.
Dutta, K., Bhushan, R., and Somayajulu, B.L.K. (2001) ΔR correction values for the Northern Indian Ocean. Radiocarbon, v.43, pp.483–485.
Dye, T. (1994) Apparent ages of marine shells: implications for archeological dating in Hawai’i. Radiocarbon, v.36, pp.51–57.
Fairbanks, R.G., Mortlock, R.A., Chiu, T.-C., Cao, L., Kaplan, A., Guilderson, T.P., Fairbanks, T.W., and Bloom, A.L. (2005) Marine Radiocarbon Calibration Curve Spanning 10,000 to 50,000 Years B.P. Based on Paired 230Th/234U/238U and 14C Dates on Pristine Corals. Quat. Sci. Rev., v.24, pp.1,781-1,796.
Fonselius, S., and Östlund, G. (1959) Natural radiocarbon measurements on surface water from the North Atlantic and the Arctic Sea. Tellus, v.11, pp.77–82.
Franke, J., Paul, A., and Schulz, M. (2008) Modelling variations of marine reservoir ages during the last 45000 years. Clim. Past, v.4, pp.125–136.
Grumet, N., Guilderson, T.P., and Dunbar, R.D. (2002) Meridional transport in the Indian Ocean traced by coral radiocarbon. Jnl. Mar. Res., v.103, pp.24,641–24,650.
Guilderson, T.P., Caldeira, K., and Duffy, P.B. (2000) Radiocarbon as diagnostic tracer in ocean and carbon cycle modeling. Glob. Biogeochem. Cyc., v.14, pp.887–902.
Higham, T.F.G., and Hogg, A.G. (1995) Radiocarbon dating of prehistoric shell from New Zealand and calculation of the ΔR value using fish otoliths. Radiocarbon, v.37, pp.409–416.
Hua, Q., Woodroffe, C.D., Barbetti, M., Smithers, S.G., Zoppi, U., and Fink, D. (2004) Marine reservoir correction for the Cocos (Keeling) Islands, Indian Ocean. Radiocarbon, v.46, pp.603–610.
Hughen, K.A., Baillie, M.G.L., Bard, E., et al. (2004) Marine04 marine radiocarbon age calibration, 0–26 cal kyr BP. Radiocarbon, v.46, pp.1,059–1,086.
Ingram, B.L. (1998) Differences in radiocarbon age between shell and charcoal from a Holocene shell mound in northern California. Quat. Res., v.49, pp.102–110.
Kalish, J.M. (1993) Pre- and post-bomb radiocarbon in fish otoliths. Earth Planet. Sci. Lett., v.114, pp.549–554.
Kalish, J.M., Nydal, R., Neadraas, K., Burr, G.S., and Eine, G.L. (2000) A 70 years long time series of radiocarbon from cod otoliths in the Barents Sea. In: Abstracts, 17th International Radiocarbon Conference, June 18–23 2000, Judean Hills, Israel, pp.94.
Libby, W.F., Anderson, E.C., and Arnold, J.R. (1949) Age determination by radiocarbon content: World-wide assay of natural radiocarbon. Science, v.109, pp.227–228.
Libby, W.F. (1955) Radiocarbon dating, 2nd edition. University of Chicago press, Chicago, U.S.A.
Linick, T.W. (1975) Uptake of Bomb-produced Carbon-14 by the Pacific Ocean. Ph.D.-Thesis, University of California, San Diego, U.S.A.
Little, E.A. (1993) Radiocarbon age calibration at archeological sites of coastal Massachusetts and vicinity. Jnl. Archeol. Sci., v.20, pp.457–471.
Madhupratap, M., Kumar, S.P., Bhattathiri, P.M.A., Kumar, M.D., Raghukumar, S., Nair, K.K.C., and Ramaiah, N. (1996) Mechanism of the biological response to the winter cooling in the northeastern Arabian Sea. Nature, v.384, pp.549–552.
Maier-Reimer, E. (1993) Geochemical cycles in an ocean general circulation model: Preindustrial tracer distribution. Glob. Biogeochem. Cyc., v.7, pp.645–678.
Milliman, J.D., and Mead, R.H. (1983) World delivery of riverine sediments to the oceans. Jnl. Geol., v.21, pp.1–21.
Mook, W.G., and van der Plicht, J. (1999) Reporting 14C activities. Radiocarbon, v.41, pp.227-239.
Moore, W.S., Krishnaswami, S., and Bhat, S.G. (1973) Radiometric determination of coral growth rates. Bull. Mar. Sci., v.23, pp.157–176.
Moore, W.S., and Krishnaswami, S. (1974) Correlation of X-radiography revealed banding in corals with radiometric growth rates. In: Proceedings, 2nd International Coral Reef Symposium, Brisbane, Australia, pp.269–276.
Nozaki, Y., Rye, D.M., Turekian, K.K., and Dodge, R.E. (1978) A 200 year record of Carbon-13 and Carbon-14 variations in a Bermuda coral. Geophys. Res. Lett., v.5, pp.825–828.
Oeschger, H., Siegenthaler, U., Schotterer, U., Gugelmann, A. (1975) A box diffusion model to study the carbon dioxide exchange in nature. Tellus, v.27, pp.168-192.
Rafter, T.A., and Ferguson, G.J. (1957) "Atomic bomb effect" — recent increase of carbon-14 content of the atmosphere and biosphere. Science, v.126, pp.557–558.
Rafter, T.A., and O'Brien, B.J. (1973) 14C measurements in the atmosphere and in the South Pacific Ocean—A recalculation of the exchange rates between the atmosphere and the ocean. In: (Rafter, T.A., and Grant-Taylor, T., eds.) Proceedings of the 8th International Conference on radiocarbon dating, The Royal Society of New Zealand, v.1, pp.241–266.
Reimer, P.J. and Reimer, R.W. (2001) A marine reservoir correction database and on-line interface. Radiocarbon, v.43, pp.461-463.
Rubin, M, and Suess, H. (1955) US Geological Survey radiocarbon dates II. Science, v.121, 481-488.
Siani, G., Paterne, M., Arnold, M., Bard, E., Bernard, M., Tisnerat, N., and Bassinot, F. (2000) Radiocarbon reservoir ages in the Mediterranean Sea and the Black Sea. Radiocarbon, v.42, pp.271–280.
Southon, J.R., Nelson, D.E., and Vogel, J.S. (1990) A record of past ocean-atmosphere radiocarbon differences from the northeast Pacific. Paleoceanogr., v.5, pp.197–206.
Southon, J.R., Nelson, D.E., and Vogel, J.S. (1992) The determination of past ocean-atmosphere radiocarbon differences. In: (Bard, E., and Broecker, W.S. eds.) The Last Deglaciation: Absolute and Radiocarbon Chronologies. Springer–Verlag, Berlin, Germany, pp.219–227.
Southon, J.R., Kashgarian, M., Fontugne, M., Metivier, B., and Yim, W.W-S. (2002) Marine reservoir corrections for the Indian Ocean and southeast Asia. Radiocarbon, v.44, pp.167-180.
Staubwasser, M., Sirocko, F., Grootes, P.M., and Erlenkeuser, H. (2002) South Asian monsoon climate change and radiocarbon in the Arabian Sea during early and mid Holocene. Paleoceanogr., v.17, 1063, doi:10.1029/2000PA000608.
Stuiver, M., and Polach, H.A. (1977) Reporting of 14C data. Radiocarbon, v.19, pp.355–363.
Stuiver, M., and Östlund H.G. (1983) GEOSECS Indian Ocean and Mediterranean radiocarbon. Radiocarbon, v.25, pp.1–29.
Stuiver, M., and Quay, P.D. (1983) Abyssal water Carbon-14 distribution and the age of the world oceans. Science, v.219, pp.849–851.      
Stuiver, M., Pearson, G.W., and Braziunas, T.F. (1986) Radiocarbon age calibration of marine samples back to 9000 cal BP. Radiocarbon, v.28, pp.980–1,021.
Stuiver, M., and Braziunas, T.F. (1993) Modeling atmospheric 14C influences and 14C-ages of marine samples to 10,000 BC. Radiocarbon, v.35, pp.137–189.
Stuiver, M., Reimer, P.J., and Braziunas, T.F. (1998) High precision radiocarbon age calibration for terrestrial and marine samples. Radiocarbon, v.40, pp.1,127–1,083.
Talma, A.S. (1990) Radiocarbon age calibration of marine shells. In: Quarterly Report, Quaternary Dating Research Unit, CSIR, Pretoria, South Africa. pp.10.
Taylor, R.E., and Berger, R. (1967) Radiocarbon content of marine shells from the Pacific coasts of Central and South America. Science, v.158, pp.1,180–1,182.
Toggweiler, J.R., Dixon, K., and Bryan K. (1989) Simulations of radiocarbon in a coarse-resolution world ocean model, 1. Steady state prebomb distributions. Jnl. Geophys. Res., v.94, pp.8,217–8,242.
von Rad, U., Schael, M., Michels K.H., Schulz, H., Berger, W.H., and Sirocko, F. (1999) A 5000-yr record of climate change in varved sediments from the oxygen minimum zone off Pakistan, northern Arabian Sea. Quat. Res., v.51, pp.39-53.
Weidman, C.R., and Jones, G.A. (1993) A shell-derived time history of bomb 14C on Georges Bank and its Labrador Sea implications. Jnl. Geophys. Res., v.98, pp.14,557–14,588.
About the author
Dr. Koushik Dutta is a scientist at AMS Radiocarbon Laboratory, Institute of Physics, Bhubaneswar. He did his Ph.D. research from Physical Research Laboratory, Ahmedabad, on studies of marine processes in the northern Indian Ocean using radiocarbon. During his tenure as postdoctoral researcher at University of Florida, Gainesville, USA, he worked on potential of carbon release from permafrost soils of Siberia and Alaska, using laboratory incubation and carbon isotopic methods. Currently his main area of research is applications of radiocarbon in earth sciences (including oceanography, palaeoclimatology, hydrology, etc.) and archaeology using accelerator mass spectrometry (AMS). His other areas of research include past variations of South Asian monsoon and oceanographic applications of 10Be and 87Sr isotopes.
 
E-mail: koushikc14(@)gmail.com  Home page: http://www.iopb.res.in/~kdutta

© The Society of Earth Scientists                                                                                                                                                                   Site Designed & Developed by SAVA Technologies