• radiocarbon;
  • methane hydrate;
  • glacial sediment;
  • foraminifera;
  • western North Pacific


  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results and Discussion
  6. 4. Conclusion
  7. Acknowledgments
  8. References

A previous study interpreted extremely 13C-depleted excursions of planktonic and benthic foraminifera in last glacial sediments (17,500 to 25,400 cal years B.P.) of the core retrieved from off Shimokita Peninsula and off Hokkaido, Japan, as evidence for periodic releases of methane, arising from the dissociation of methane hydrate. To better understand the formation process of the 13C-depleted excursions, we conducted high-resolution natural radiocarbon measurements and biogeochemical analyses. We found highly depleted 13C excursions ranging from −10.2‰ to −1.6‰ and −6.8‰ to −1.6‰ in planktonic and benthic foraminifera, respectively. Most of the foraminiferal tests in these horizons were brown, most likely as a result of postdepositional alteration, reflecting the formation of authigenic carbonate on the surface of tests. These alterations were also supported by high levels of Mg-calcite and the acid-leaching test for anomalous foraminifera. To evaluate the carbon sources in the altered foraminifera tests, we quantified the relative contributions of 14C-free methane-derived carbon sources to the formation of authigenic carbonates in foraminifera with depleted 13C excursions using a coupled mass balance isotopic model (14C/C and 13C/12C). The radiocarbon ages of both planktonic and benthic 13C-depleted foraminifera were approximately 600 to 2000 years older than those of normal tests from nearby horizons. The relative contributions of authigenic carbonates derived from the methane oxidizing process reached to ∼22 wt% for planktonic foraminifera and ∼15 wt% for benthic foraminifera. The δ13C values of methane calculated from the mass balance model were between −29‰ and −68‰ for planktonic foraminifera and between −40‰ and −108‰ for benthic foraminifera, consistent with δ13C values reported for thermogenic and abiogenic methane in global methane hydrate reservoirs. These data consistently suggest that methane-related drastic environmental change occurred in the horizons that included δ13C anomalies. This study provides important information for interpreting geological records of the methane hydrate instability associated with climate.

1. Introduction

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results and Discussion
  6. 4. Conclusion
  7. Acknowledgments
  8. References

Evidence of methane hydrate instability has been reported from marine and lake sediments of the Quaternary [e.g., Wefer et al., 1994], the Late Paleocene [e.g., Dickens et al., 1995, 1997], Early Cretaceous [Jahren et al., 2001], and the Jurassic [Hasselbo et al., 2000]. Their evidence for Late Quaternary hydrate instability is reported in marginal sea sediments such as the California Margin [Hinrichs et al., 2003; Kennett et al., 2000], Gulf of California [Keigwin, 2002], Papua New Guinea Margin [de Garidel-Thoron et al., 2004], Amazon Fan [Maslin et al., 1998], and Greenland Sea Margin [Millo et al., 2005a; Smith et al., 2001], and in lake sediments in Lake Baikal [Prokopenko and Williams, 2004, 2005]. Kennett et al. [2000] suggested, on the basis of foraminiferal δ13C in the Santa Barbara basin (SBB), the “clathrate gun” hypothesis, which highlights the potential role of methane hydrate in Quaternary variations in atmospheric methane concentrations. Moreover, biomarker records supporting methane hydrate instability were documented in the SBB [Hinrichs et al., 2003] and off the Shimokita Peninsula, Honshu island, Japan [Uchida et al., 2004]. These biomarker records provided aerobic methanotrophic molecular fossil evidence for periodic methane release, which may also support the clathrate gun hypothesis. More recently climate-sensitive hydrocarbon seepage of oil and gas, including large amounts of methane, has also been reported from the SBB during the last deglaciation [Hill et al., 2006]. The extent of support that foraminiferal δ13C records from sediments lend to the clathrate gun hypothesis are mainly dependent, however, on whether the planktonic δ13C excursions represent primary seawater δ13C values or secondary, postdepositional diagenetic δ13C values. Regarding interpretation of these foraminiferal δ13C records, some studies seem to be questionable, especially about the extent of methane oxidation that occurs in the water column. Recent studies have proposed that contamination by authigenic overgrowths or calcite replacement may be difficult to detect through microscopy [Hill et al., 2004a, 2004b]. Acid incremental leaching experiments for 13C-depleted foraminifera have been examined in detail to quantify the extent of authigenic overgrowth [Millo et al., 2005b; Torres et al., 2005]. Additionally, organic carbon in situ could contaminate the foraminiferal isotopic records, leading to a more depleted 13C composition, because total organic carbon (TOC) δ13C values (∼−23‰ to −22‰) are much lower than foraminiferal δ13C values (0‰ to −1‰). It is a matter of ongoing debate whether such extremely negative foraminiferal δ13C values are derived from dissolved inorganic carbon (DIC) at primary calcification or from subsequent authigenesis [Cannariato and Stott, 2004]. To address this controversy, several researchers have conducted extensive studies that have reached conflicting and complex conclusions. Thus we need to know the correct mechanism explaining the extremely 13C-depleted foraminifera found in glacial/interglacial sediments.

Ten or more regions from the northwest Pacific along the Japanese islands (including the Oyashio current region, the Sea of Okhotsk, the Nankai Trough, and the Japan Sea) have been identified as immense methane hydrate reservoirs [Center for Deep Earth Exploration (CDEX), 2002; Kvenvolden and Lorenson, 2001; Satoh, 1994, 2002] (Figure 1). Despite the potential for large episodic releases of methane in the western North Pacific, the relationship between climate and hydrate instability in the region has not been studied sufficiently. In a previous study that used the same core as was used in this study (core PC6, off Tokachi, Hokkaido, northeast Japan), many extremely 13C-depleted foraminiferal signals were found from sediment layers ranging in age from 21,000 to 17,800 cal years B.P. [Ohkushi et al., 2005]. Recent observations have shown the existence of a bottom-simulating reflector (BSR) over a large area that includes this study site [Satoh, 2002; TuZino and Noda, 2007].


Figure 1. (a) Location map of core sites PC6 (this study site), PC4/5, and CK. The red dot indicates the locations of the epicenters of the Tokachi-oki interplate earthquakes of 1952 and 2003 (41.780°N, 144.079°E; depth, 42 km; moment magnitude 8.0; 26 September 2003). (b) Detailed location map of core site PC6. The black dots indicate the locations of the epicenters of the Tokachi-oki interplate earthquakes of 1952 and 2003 (41.780°N, 144.079°E; depth, 42 km; moment magnitude 8.0; 26 September 2003). The yellow shaded areas show the locations of anomalous BSRs, which suggest the presence of methane gas hydrates [TuZino and Noda, 2007]. Each location has been described in detail previously by Kvenvolden and Lorenson [2001], Satoh [1994, 2002], and Satoh et al. [1994]. The 13C-depleted planktonic and benthic foraminifera for several horizons in the PC4/5 cores were reported by Uchida et al. [2004] and Hoshiba et al. [2006]. CK C9001 sites (41.10°N, 142.12°E; depth, 1180 m) represent the location at which methane hydrate was recently collected from the core depth of 189 m under the seafloor at a water depth of 1180 m during expeditions by the CK06-06 leg1 cruise by R/V Chikyu (information available at [Taira and Curewitz, 2005].

Download figure to PowerPoint

In this study, we performed high-resolution natural radiocarbon measurements of planktonic and benthic foraminifera and organic carbon to decipher the carbon source of foraminiferal carbon isotope anomalies. 14C measurements of fossil foraminiferal carbonates were an essential tool for correctly determining the ages of the sediment horizons, because its 5730-year half-life of 14C makes it a convenient tracer of processes that occur on a millennial timescale (less than about 50,000 years). Using obtained natural radiocarbon data of planktonic and benthic foraminifera, we quantitatively evaluated the carbon source of foraminiferal isotopic anomalies by a coupled isotopic mass balance model (14C/C and 13C/12C).

Foraminifera use the DIC in the surrounding seawater when making their carbonate skeletons; thus the 14C content of the foraminiferal carbonate should reflect the 14C content of the DIC of the seawater when the foraminifera were alive. Methane hydrates buried under the deep seafloor have been known as fossil carbon (14C-free) because the methane hydrate was biogenically formed using ancient organic carbon or originated through thermodecomposition of fossil organic carbon [Grabowski et al., 2004; Kessler et al., 2005, 2006; Winckler et al., 2002]. Thus, if methane hydrate dissociation occurred as a result of any environmental changes and was accompanied by the oxidation of methane, the 14C content of seawater would be highly contaminated with the 14C-free DIC derived from methane. In this study, we took advantage of the complete absence of 14C in methane hydrates to quantify the carbon source derived from methane hydrate dissociation recorded in the foraminiferal δ13C anomalies. We also used analyses of trace metals, such as the Mg/Ca ratio, as an index of authigenic overgrowth and a phylogenetic type analysis to evaluate the past activity of methanotrophs and methanogens associated with past drastic environmental changes.

2. Material and Methods

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results and Discussion
  6. 4. Conclusion
  7. Acknowledgments
  8. References

A variety of geochemical techniques were used in this study, including 13C, 14C, 18O isotopic signatures of foraminiferal fossils and sedimentary organic matter and trace metal (Mg/Ca, Mn/Ca ratio). Carbon isotopic data, especially 14C, were used for the first time to determine the relative contributions of carbon to authigenic carbonate formation and/or biogenic calcite using the oxidation of methane and/or methane hydrate. Additionally, to make the puzzles of various geochemical data fit together more easily, we used phylogenetic analysis and quantitative PCR analysis of archaeal 16S rRNA genes, which may provide information on the past microbial activity with respect to methane.

2.1. Sampling Location and Core Lithology

The piston core (MR01-K03 PC6, 42° 21.42′N, 144° 13.36′E; Figure 1 and Figure 2) was retrieved from a single site (a water depth 1066-m) in the eastern margin of Hokkaido, northern Japan in the northwest Pacific during MR01-K03 cruise of R/V Mirai. The sediment consists of massive, olive gray, bioturbated, diatom-bearing sandy silts. Foraminifera, shell fragments, parts of plant fossils, and white pumice grains are scattered sporadically throughout the core. The benthic foraminiferal assemblage is dominated by bathyal species such as Uvigerina akitaensis and Elphidium batialis. The planktonic foraminifera consist of subarctic taxa such as Neogloboquadrina pachyderma (sinistral). The odor of H2S is prominent in the middle to lower part of the core. Burrows are often filled with pyritic grains. Turbidite layers occur at 36–47 cm, 180–182 cm, and 211–214 cm depth in the core. Thin volcanic ash layers are intercalated at 225–226 cm and at 284 cm.


Figure 2. Seismic profile transecting the Tokachi forearc basin. A BSR underlies the seafloor at a depth of 0.4–0.5 s TWT (∼300–375 mbsf if the velocity is 1.5 km s−1). The hand signs show pipe structures. Q, Quaternary; P, Pliocene; M, upper Miocene.

Download figure to PowerPoint

2.2. The 14C Analysis of Foraminiferal Tests and Sedimentary Organic Carbon

AMS 14C dates are used to build the chronology of core PC6 and described in detail by Ohkushi et al. [2005] according to AMS14C dating on planktonic foraminifera picked from 14 horizons. The new more data of 45 samples of planktonic and benthic foraminifera including 13C-depleted specimens and 18 samples of sedimentary organic carbon were obtained in this study. The results are listed in Table 1. Samples were washed with tap water over a 63 μm screen, rinsed in distilled water and then dried in 50°C. Using a binocular microscope, planktonic foraminifers were hand-picked from samples, and then cleaned by soaking in 30% hydrogen peroxide solution to remove adhering contaminants. If enough abundance was obtained, single species (N. pachyderma) was used for planktonic dating. Sedimentary organic carbon was pretreated with 1N HCl solution to remove carbonate in the sediment sample before graphitization.

Table 1. Elemental Data and Isotopic Signatures of Planktonic and Benthic Foraminifera and Sedimentary Organic Carbon From the Horizons Including Anomalous Foraminiferal Testsa
Sample IDDepth in Core, cmU. akitaensisbδ13C, ‰N. pachydermabCalendar Age, years B.P.U. akitaensisN. pachydermaTotal Organic Carbon (TOC)
Anomaly Eventδ18O, ‰δ18O, ‰δ13C, ‰Age, years B.P.Δ14C, ‰Lab. CodeAge, years B.P.Δ14C, ‰Lab. CodeTOC, %TOC δ13C, ‰Age, years B.P.Δ14C, ‰Lab. Code
  • a

    Elemental data are total organic carbon; the isotopic signatures used are 13C/12C, 18O/16O, and 14C.

  • b

    All δ18O and δ13C of planktonic and benthic foraminifera and part of the radiocarbon data of planktonic foraminifera data are derived from Ohkushi et al. [2005].

  • c

    These 14C measurements were carried out in NOSAMS, Woods Hole Oceanographic Institution. Other 14C data were acquired in AMS facility (NIES-TERRA), National Institute for Environmental Studies.

1-1430 4.09−0.68  9887   9860 ± 70−707 ± 6TERRA-121503a200.91−22.3   
1-2453     13609   13420 ± 80−812 ± 8TERRA-121503a23     
1-3475     15744   14020 ± 80−825 ± 8TERRA-121503a24 −23.7   
2-8100 4.7−13.890.0616134   14320 ± 80−832 ± 8TERRA-121503a250.78−24.1   
2-20128     16897   14960 ± 140−845 ± 15TERRA-121503a26     
3-2192     17156   15200 ± 70c−849 ± 7OS-34590  16990 ± 80−879 ± 9TERRA-071404a12
3-14222              16460 ± 100−871 ± 11TERRA-063006a01
3-20234 4.78−1.533.860.081750615780 ± 80−860 ± 8TERRA-091305a0514680 ± 80−839 ± 8TERRA-080604a15     
3-212361a4.77−3.094.55−7.671752516250 ± 80−868 ± 9TERRA-091305a1516650 ± 100−874 ± 11TERRA-081605a190.67−23.317070 ± 90−881 ± 10TERRA-063006a02
3-232411a4.86−6.774.47−10.171756217040 ± 130−880 ± 14TERRA-091305a1316710 ± 90−875 ± 10TERRA-081605a200.67−23.417720 ± 90−890 ± 10TERRA-063006a03
3-242431a4.85−4.124.09−4.701758116690 ± 170−875 ± 19TERRA-091305a1416520 ± 110−872 ± 12TERRA-081605a21     
3-25245 4.73−1.253.830.121760016200 ± 100−867 ± 11TERRA-091305a0915380 ± 100−853 ± 10TERRA-091305a16     
3-322611b4.72−1.323.77−1.561773016030 ± 290−859 ± 11TERRA-091305a0615750 ± 100−864 ± 31TERRA-121503a300.63−24.016920 ± 100−878 ± 11TERRA-063006a04
3-332631b5.31−5.854.36−8.081773116620 ± 80−874 ± 9TERRA-091305a1916580 ± 80−873 ± 8TERRA-042505a260.71−23.617830 ± 90−891 ± 10TERRA-063006a05
3-342651b4.76−2.704.19−4.7517733   16490 ± 110−872 ± 12TERRA-072805a14     
3-36269 4.75−1.253.820.141773515960 ± 80−863 ± 9TERRA-091305a2616200 ± 320−867 ± 35TERRA-072805a13     
3-38274 4.74−1.303.660.0417738   15760 ± 130−859 ± 14TERRA-072805a12     
4-8311 4.74−1.133.790.001776016260 ± 90−868 ± 10TERRA-091305a0315720 ± 150−859 ± 16TERRA-121503a290.68−23.516410 ± 110−870 ± 11TERRA-063006a06
4-9313 4.81−1.20  1778616190 ± 80−867 ± 8TERRA-091305a0716190 ± 80−857 ± 13TERRA-091305a10     
4-103151c4.81−1.643.71−1.641781316820 ± 90−877 ± 10TERRA-081605a2416490 ± 90−872 ± 10TERRA-091305a18     
4-113181c4.97−4.184.23−5.651783916570 ± 80−873 ± 8TERRA-091305a0916700 ± 90−875 ± 10TERRA-081605a220.53−23.518320 ± 100−898 ± 11TERRA-063006a07
4-123201c4.88−4.044.13−4.471786616790 ± 90−872 ± 9TERRA-091305a11        
4-133221c4.84−1.773.86−1.071789316310 ± 80−869 ± 8TERRA-091305a0416300 ± 90−869 ± 10TERRA-081605a23     
4-14324 4.75−1.243.760.131791916920 ± 90−878 ± 10TERRA-081605a25        
4-15326     1793716380 ± 80−870 ± 8TERRA-091305a1216040 ± 90−864 ± 10TERRA-091305a17     
4-38377 5.04−1.174.12−0.4418908   16000 ± 140−864 ± 15TERRA-080604a170.48−23.518380 ± 90−899 ± 11TERRA-071404a13
5-2400     19330   17100 ± 60c−881 ± 7OS-345910.43 19120 ± 110−908 ± 13TERRA-071404a14
5-32469 5.26−0.823.900.0520299   17939 ± 100−893 ± 11TERRA-121503a340.51−23.6   
6-4506     20657   18120 ± 180−895 ± 20TERRA-112905a05  19690 ± 100−914 ± 12TERRA-071404a15
6-10520              19950 ± 160−917 ± 18TERRA-071404a16
6-15530 5.230.72  20889   18450 ± 75c−900 ± 8OS-345920.48−23.5   
6-30560              20740 ± 120−924 ± 13TERRA-071404a19
6-38584 4.83−1.003.75−0.292147420000 ± 100−917 ± 12TERRA-081605a2618920 ± 90−905 ± 10TERRA-091305a25     
6-3958635.23−2.194.20−5.402149720100 ± 100−918 ± 12TERRA-081605a2719390 ± 100−907 ± 11TERRA-091305a240.52−23.020860 ± 130−925 ± 14TERRA-063006a09
6-40588     2152119880 ± 100−916 ± 11TERRA-091305a2319220 ± 100−910 ± 11TERRA-091305a21     
6-41590 4.73−1.28  2154419630 ± 100−913 ± 11TERRA-091305a22        
6-42592     2156719530 ± 110−912 ± 13TERRA-091305a08        
7-6614     21798   19240 ± 110−910 ± 13TERRA-121503a33     
7-22651     22206   19430 ± 190−911 ± 22TERRA-082405a12     
7-32               19110 ± 100−907 ± 11TERRA-071404a22
7-42696     22698   20600 ± 80c−923 ± 8OS-34593  21220 ± 110−929 ± 13TERRA-071404a20

Most of 14C data were acquired at NIES-TERRA AMS facility, National Institute for Environmental Studies [Tanaka et al., 2000; Yoneda et al., 2004]. Only four 14C measurements for planktonic foraminifera were conducted at NOSAMS, Woods Hole Oceanographic Institution. Standard sample preparation methods in NOSAMS were used [e.g., McNichol et al., 1994]. For preparation of sedimentary organic carbon before graphitization, all carbonate in the sediment were dissolved with 1N HCl solution for 1 night and then washed with mill Q water. The graphitization of foraminifera and sedimentary organic carbon in NIES-TERRA is carried out according to a procedure by Uchida et al. [2004, 2005].

To construct the age scale, we used a ΔR value of 376 years for the reservoir age correction of our planktonic 14C ages, following [Duplessy et al., 1989]. Calibrated ages were calculated against INTCAL98 [Stuiver et al., 1998] using CALIB4.3 software and the equation of Bard [1998]. Planktonic foraminifera-based calendar ages show that the core recorded paleoenvironmental changes during the past 22.6 ka. The linear sedimentation rate (LSR) was calculated from the radiocarbon data. The LSR during the last glacial period is generally high with about 33–187 cm/ka [Ohkushi et al., 2005].

2.3. Stable Carbon and Oxygen Isotopes of Foraminiferal Tests and Sedimentary Organic Carbon

Planktonic (N. pachyderma [sinistral]; 180–212 μm) and benthic (U. akitaensis; 212–500 μm) foraminifera from PC6 core samples were mainly used for the stable carbon and oxygen isotope analyses in this study. 25 specimens for N. pachyderma and 2 or 3 specimens for U. akitaensis were used for each isotopic measurement. Additionally, benthic foraminifera Elphidium batialis (250–500 μm), Nonionellina labradorica (250–500 μm), and Globobulimina auriculata (300–600 μm), and planktonic foraminifera Globigerina bulloides (250–300 μm) were measured only in isotope anomaly horizons 1a-1c (230–330 cm depths). 2 or 3 specimens for E. batialis, 2 or 3 specimens for N. labradorica, 1 or 2 specimens for G. auriculata, and 10 specimens for G. bulloides were used for each isotopic measurement. Samples 2.2 cm thick were removed from the core about every 4.4 cm. Oxygen and carbon isotopic compositions in foraminiferal tests were measured on a Finnigan MAT252 fitted with an automated carbonate preparation device (Kiel III) at the Mutsu Institute for Oceanography, JAMSTEC. Sample gases were calibrated using NIST-18 and NIST-19 reference samples, and all results are expressed in the δ-notation as per mil deviation from VPDB, using standard procedures. The analytical precision (σ) of the system was better than 0.07‰ for oxygen and 0.04‰ for carbon during this study.

Before the δ13C measurements of organic carbon in the sediments, the inorganic carbon were removed using 1N HCl. δ13C of sedimentary organic carbon in the sediments horizons corresponding to foraminiferal analyses of 14C and 13C and 18O were performed by elemental analyzer–continuous flow isotope ratio mass spectrometry (EA/CFIRMS), using a Carlo Erba NC2500 interfaced through a Finnigan CONFLOII to a Finnegan Delta XL mass spectrometer.

In the acid leaching experiment, stable carbon and oxygen isotopic compositions of single foraminiferal specimen of U. akitaensis and N. pachyderma (13C-depleted foraminiferal tests from 266–268 cm core depth), are measured using the analytical system for sub-microgram quantities of CaCO3 at Hokkaido University [Ishimura et al., 2004]. Through the analysis, some individuals are washed by weak acid (0.01N HCl) for 20 min to remove the secondary addition of authigenic carbonate. The system consists of a micro volume CaCO3 decomposition tube, stainless steel CO2 purification vacuum line with an introduction quantity-regulating unit, helium-purged CO2 purification line, gas chromatograph, and continuous-flow isotope ratio mass spectrometry (HP6890 with Finnigan MAT252).

All isotope ratios are expressed in δ notation, or parts per thousand deviation from the Vienna Pee Dee Belemnite (VPDB) standard for 13C/12C and VPDB and SMOW standards for 18O/16O, where

  • equation image
  • equation image

2.4. Trace Metal Analysis for Foraminiferal Tests

Benthic foraminifera U. akitaensis was hand picked from the coarse fraction, and >300 μm fraction was used for Mg/Ca and Mn/Ca analysis. 10 specimens were used for a single analysis. Solutions were prepared from H2O (deionized and single distilled in a quartz chamber) and ultra pure 15M HNO3 (Tamapure AA100®, Tama Chemical). All plastic wares (Teflon® and polypropylene bottles and beakers) were leached in 7M HNO3 at room temperature for 5 days and then rinsed with Milli-Q (18.3 MΩ) water, and were dried.

All chemical cleaning processes were carried out in the laminar flow bench in a class 10,000 clean room. Preparation of fossil Mg/Ca determination followed the procedures of trace metal analysis of foraminifera [Boyle and Keigwin, 1985; Rosenthal et al., 1999]. Individual foraminifers were slightly crushed to open the each chamber. After that, fragments were moved to a micro tube and were purified by an ultrasonic agitation, reduction, and oxidation process. Finally, the samples were leached several times by weak acid and were completely dissolved by 0.3 M ultra pure HNO3.

For determination of Mg/Ca ratio, we used the magnetic sector field inductively coupled plasma mass spectrometry (Thermo Finnigan, ELEMENT2). The measurement was carried out in medium resolution mode (m/Δm = 4,000). The sample introduction system consisted of a MicroFlow PFA-100 self-aspirating Teflon nebulizer (Elemental Scientific Inc, aspiration rate is 100 μl/min). Standard elemental solutions used were the SPEX Claritas PPT® certified solutions. To all samples and standard solutions was added 1 ppb Sc as an internal standard. Mn/Ca was also routinely monitored for evaluating the secondary overgrowths of MnCO3 by sedimentary diagenesis.

2.5. The 16S rRNA Gene Analysis of Archaeal Community Structure

Bulk DNA was extracted from 10 g of wet sediments with the Soil DNA Mega Prep kit (Mo Bio Lab., Inc., Solana Beach, CA, USA) and then purified and concentrated as previously described [Inagaki et al., 2003]. Archaeal 16S rRNA genes were amplified by PCR using Arch 21F and Arch 958R primers [DeLong, 1992]. The PCR conditions were as follows: 30 s denaturation at 96°C, 30 s annealing at 50°C, and 120 s extension at 72°C for the amplification of 35 cycles. The archaeal 16S rRNA genes were cloned and sequences as previously described [Inagaki et al., 2003, 2006]. For the phylogenetic analysis of archaeal community structures, approximately 450 nucleotides of all 16S rRNA clones were sequenced with Arch 21F primer. Similarity among clone sequences was analyzed using FASTA program with DNASIS software (Hitachi Co., Tokyo, Japan). The sequences having >97% similarity were assigned to phylogenetic representatives. The representative sequences were subjected to similarity analysis using the FASTA 3 and the gapped-BLAST search algorithms against GenBank/EMBL/DDBJ databases.

2.6. Seismic Profile Around This Study Site

Seismic data were collected by a six-channel seismic system in the course of the research cruise GH02 (June, 2002) of the R/V Hakurei Maru No. 2. The sound source of the seismic profiling systems consisted of a 355 cu. in. G.I. gun (Sercel, France), operated at 20–100 Hz. The profiles have a vertical resolution of 0.03 s two-way travel time (TWT), the penetration varying between 1 and 2 s TWT. The strata were correlated with the boring core of the MITI Tokachi-oki [Sasaki et al., 1985; TuZino and Noda, 2007].

3. Results and Discussion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results and Discussion
  6. 4. Conclusion
  7. Acknowledgments
  8. References

3.1. Distribution of BSR Around the Study Site

Figure 2 shows the seismic reflection profile of this study site (PC6). The core collection point is represented as an array in the Figure 2. A BSR is widely distributed in the Tokachi forearc basin [Satoh, 1994, 2002; TuZino and Noda, 2007]. The PC6 core site is at the top of an anticline composed of Miocene-Pliocene forearc basin fill and also located above the BSR (Figure 2).

The BSR cuts the stratal reflectors. The polarity of the BSR is inverted in relation to the reflectors of the seafloor and strata. The BSR enhances the underlying reflectors and shows the alignment of the chaotic reflection patterns within the anticline. These features suggest that the BSR represents the upper boundary of a free gas layer. The overlying strata on the anticline are characterized by numerous pipe structures or discontinuities of the reflections (the hands shown in Figure 2). This strata may also imply an extensional stress field on the anticline because of the bended flexure. The anticline was produced by regional compression, which can drive a rise of fluid. Thus the architecture might be suitable for gas seepage. The top of the anticline can be interpreted as a depressed structure, and the Miocene strata show scattered reflections. Those reflections might be related to gas seepage, although the uppermost reflectors show neither pockmarks nor diapirs.

3.2. Foraminiferal Isotopic Anomalies in the LGM-Deglaciation Transition Ages

The δ13C values for benthic and planktonic foraminifera over the past 22.6 ka are shown in Figure 3. A detailed description of the foraminiferal isotopic signatures for δ18O and δ13C is also reported by Ohkushi et al. [2005]. Normal benthic U. akitaensisδ13C values during the glacial period were stable, ranging from −1.0‰ to −0.5‰ (Figure 3a and Table 1). However, the benthic δ13C record during the last glacial maximum (LGM) was punctuated by six negative anomalies (1a–4) of about 1‰ to 5‰, from 241 cm to 367 cm and at 586 cm and 687 cm. The values for U. akitaensis decrease to −6.77‰ (Δ = −6‰ from the background) at anomaly 1a, −5.85‰ (Δ = −5‰) at anomaly 1b, and −4.18‰ (Δ = −3‰) at anomaly 1c. Glacial planktonic δ13C values of N. pachyderma had normal values of about 0‰, but the planktonic δ13C record was also punctuated by at least four highly negative anomalies (1a–2) of about 3‰ to 10‰. The values for N. pachyderma decrease to −10.44‰ (Δ = −10‰ from the background) at anomaly 1a, −8.08‰ (Δ = −8‰) at anomaly 1b, and −5.65‰ (Δ = −5‰) at anomaly 1c. Sedimentary ages of the horizons with the three largest anomalies (1a, 1b, and 1c), observed at 241, 263, and 318 cm, were estimated as 17.6, 17.7, and 17.8 cal kyr B.P., respectively. The sedimentary ages of horizons with anomalies 2, 3, and 4, at 367, 586, and 687 cm, were 18.8, 21.5, and 22.6 cal kyr B.P., respectively.


Figure 3. (a) Benthic (U. akitaensis) and (b) planktonic (N. pachyderma) foraminiferal δ13C records from PC6 by depth in the core (derived from Ohkushi et al. [2005]). Numbers from 1a to 4 indicate negative δ13C anomaly events.

Download figure to PowerPoint

3.3. Geochemical Signatures and Elemental Analysis

The total organic carbon (TOC) and their δ13Corg values are shown in Figure 4 and Table 1. Values of TOC and δ13Corg from the sediment interval spanning the anomaly events (1a–1c, 2, and 3) range from 0.43% to 0.91% and −22.3‰ to −24.1‰, respectively. The TOC decreases progressively with core depth to 400 cm, and values at core depths between 400 cm and 700 cm were constant at about 0.48%. In contrast, δ13Corg values were most enriched in 13C at the core surface and constant in all other horizons. The TOC and δ13Corg values from the upper horizon of the core in this study were consistent with those from surface sediments on the shelf and the slope near this study area [Nagao et al., 2005]. In situ microbial oxidation of organic carbon may contaminate foraminiferal isotopic records leading to a more depleted 13C composition because TOC δ13C values (∼ −23‰ to −2‰) are much lower than foraminiferal δ13C values (0‰ to −1‰). It is a matter of ongoing debate whether such extremely negative foraminiferal δ13C values are derived from the DIC composition at primary calcification or from subsequent authigenesis associated with in situ oxidation of organic matter [Cannariato and Stott, 2004]. The clathrate dissociation scenario proposed to explain the negative δ13C excursions recorded by foraminifera requires the transfer of 13C-depleted carbon from clathrate methane to dissolve CO2 in the water column [Kennett et al., 2000]. Phytoplankton growing in surface waters should record a similar δ13C excursion in TOC. High-resolution records of TOC δ13C values in the sediment horizon across the foraminiferal δ13C excursion from MIS 3 in Santa Barbara [Hinrichs et al., 2003] exhibited variations similar in magnitude and timing to the TOC δ13C values that occurred in the basin over the last 600 years [Schimmelmann, 1991]. In this study, similar variations of TOC δ13C and foraminiferal anomalies are not clear (Figure 5).


Figure 4. Depth profiles of (a) total organic carbon (TOC) (%) and (b) δ13Corg. The arrays in Figures 4a and 4b correspond to the horizons that include the foraminiferal anomalies listed in Table 1.

Download figure to PowerPoint


Figure 5. (a) Relationships between total organic carbon (TOC) (%) and δ13C values of normal and anomalous foraminifera. (b) Relationships between δ13Corg and δ13C values of normal and anomalous foraminifera.

Download figure to PowerPoint

3.4. Incremental-Leaching Experiments of Foraminiferal Tests

The results of acid incremental-leaching experiments of δ13C anomaly benthic and planktonic foraminiferal tests are shown in Figure 6. Average normal δ18O and δ13C values are 4.8‰ and −1.2‰ for benthic foraminifera and 3.8‰ and 0‰ for planktonic foraminifera, reflecting the δ13C and δ18O values of primary DICseawater. The δ18O values of δ13C anomaly foraminiferal tests are isotopically heavier by 0.6‰ of planktonic foraminifera and by 0.3‰ of benthic foraminifera than those of normal species. This trend is consistent with those in authigenic carbonate [Formolo et al., 2004] and carbonate fossils in a modern methane seep environment [Torres et al., 2003]. The arrows (A, C) in Figure 6 show transitions of δ13C and δ18O values of anomaly benthic and planktonic foraminiferal tests after incremental-leaching treatment. On the other hand, the arrows (B, D) represent transitions in which case an authigenic overgrowth is only contributor to account for negative δ13C excursions. From the results of leaching tests, the δ13C and δ18O values of anomaly planktonic and benthic foraminiferal tests moved to different values with more depleted 13C value (ΔPF = 8‰ and ΔBF = 5‰) and more enriched 18O value (ΔPF = 0.5‰ and ΔBF = 0.2‰) compared with normal values, as shown by the arrows (A, C). These results strongly imply mixtures of various carbon sources derived from primary calcification (biogenic), as well as secondary overgrowth (post-depositional authigenic), with respect to the formation of the anomalous foraminifera carbonate found at this study site.


Figure 6. Stable carbon and oxygen isotopic records of 13C anomalous planktonic (N. pachyderma) and benthic (U. akitaensis) foraminiferal species following acid-leaching treatments of the surface of foraminiferal tests. Circles and squares represent planktonic and benthic foraminifera, respectively. Solid and open symbols represent water-washed and acid treatment, respectively. Open squares and circles with dots inside represent the normal values of planktonic and benthic foraminifera collected from nearby anomalous horizons.

Download figure to PowerPoint

3.5. Mg/Ca and Mn/Ca Ratios of Foraminiferal Tests

We measured Mg/Ca and Mn/Ca ratios for U. akitaensis in the δ13C anomalous and normal horizons to evaluate the influence of authigenic overgrowth on the foraminiferal tests. The results are given in Table 2, along with δ18O and δ13C values. Typically the foraminiferal Mg/Ca ratio is used to reconstruct paleo-sea surface temperature (SST) [Elderfield and Ganssen, 2000; Lea et al., 1999; Nurnberg et al., 1996]; values generally range from 1.0 mmol/mol to 1.2 mmol/mol, which is a small variation. All normal foraminifera had values of about 1.0 mmol/mol, but the results were highly variable for the anomalous foraminifera: the maximum value was as high as 7.3 mmol/mol at anomaly event 1a, which had a δ13C value of −5.85‰. All of these measurements were carried out following Rosenthal et al.'s [1999] typical cleaning procedure, which is described in detail in section 2.3. Such high Mg/Ca values have also been reported from deep-sea sediments that contain foraminifera that are highly 13C depleted [de Garidel-Thoron et al., 2004] as well as from cold seep environments [Torres et al., 2003], suggesting that production of authigenic calcite occurred. The Mg/Ca ratios for anomalous foraminifera in this study are consistent with reported values (2.6–3.3 mmol/mol) from the glacial horizons in the Gulf of Papua [de Garidel-Thoron et al., 2004]. The high Mg/Ca ratios suggest authigenic overgrowth of foraminifera associated with a drastic change in the concentration of trace metals in the water column. Such a change may be caused by methane hydrate dissociation. Foraminiferal Mn/Ca ratios are also shown in Table 2. The Mn/Ca ratios show similar patterns as the Mg/Ca ratios. The Mn/Ca ratios are also used as another index to evaluate the influence of overgrowth on foraminiferal tests. The Mg/Ca and Mn/Ca ratios may indicate the possibility of both biogenic calcification in the water column and authigenic calcification in pore water and/or sediments. This result is also likely supported by the incremental-leaching experiment of anomalous foraminifera.

Table 2. Isotopic Values, Mg/Ca Ratio, and Mn/Ca Ratio for Benthic Foraminifera Species Analyzed in the PC6 Core, MR01-K03a
Depth in Core, cmAnomaly Eventδ18O, ‰ VPDBδ13C, ‰ VPDBMg/Ca, mmol/molMn/Ca, μmol/mol
  • a

    The benthic foraminifera species is U. akitaensis.


Other evidence of authigenic overgrowth was clear from the microscope images, shown in Figure 7. The pictures are of benthic foraminifera from the 1a anomaly horizon and from a glacial normal horizon. Figure 7 also shows a comparison of the weight between normal and anomalous tests based on their carbon content measured as CO2. The weight of anomalous tests are clearly heavier than that of normal tests. The carbon content of the altered foraminiferal tests decreased progressively with the magnitude of the 13C excursion, by as much as 70% when compared with normal foraminiferal tests. These results indicate that part of the altered tests may be composed of other trace metals and clay minerals. Additionally, the average weight of the anomalous foraminiferal specimens was significantly larger than that of the normal foraminifera. The average weights were about 4.6 μg for anomalous planktonic (N. pachyderma) foraminifera (241 cm depth) versus about 2.8 μg for normal planktonic foraminifera (245 cm depth) 212–300 μm in size. The averages were about 43 μg for anomalous benthic (U. akitaensis) foraminifera (241 cm depth) versus about 27 μg for normal benthic foraminifera (245 cm depth) 300–450 μm in size.


Figure 7. Relationships between the relative weight percent of altered foraminifera versus unaltered foraminifera and the δ13C values of foraminifera. The inset photos show both normal white and abnormal brown colored specimens of the benthic foraminifer U. Akitaensis (test diameter: about 1 mm).

Download figure to PowerPoint

3.6. The 14C and 13C Isotopic Signatures of Foraminiferal Tests and Sedimentary Organic Carbon

Δ14C and δ13C records of planktonic and benthic foraminifera in δ13C anomaly horizons 1a–1c in the core are shown in Figure 8 and Table 1. Δ14C values were analyzed for planktonic (N. pachyderma) and benthic (U. akitaensis) foraminifera. 14C ages of planktonic and benthic foraminifera and sedimentary OC (TOC) collected from the anomalous and normal horizons are shown in Figure 9. All Δ14C values of anomalous foraminifera were depleted in 14C as compared with normal foraminifera from the horizons spanning the anomalous foraminifera (Figure 8a). The Δ14C differences between normal and anomalous tests ranged from 5‰ to 35‰ for planktonic foraminifera and from 3‰ to 20‰ for benthic foraminifera. The maximum corresponding age differences were found for the 1a event, about 2000 years for planktonic foraminifera and 1300 years for benthic foraminifera. The corresponding δ13C values were also the most 13C-depleted values, −10‰ for planktonic foraminifera and −7‰ for benthic foraminifera. Other foraminifera species in the other anomalous events had also similar patterns (Figures 8b and 8c). The planktonic foraminifer G. bulloides exhibited δ13C values of −7.54‰ (Δ = ∼−7‰ from the background) at anomaly 1a, 8.62‰ (Δ = ∼−8‰) at anomaly 1b, and −6.09‰ (Δ = ∼−5‰) at anomaly 1c. Likewise, the benthic foraminifer E. batialis values decreased to −7.47‰ (Δ = ∼−4‰) at anomaly 1a, −8.62‰ (Δ = ∼−6‰) at anomaly 1b, and −7.14‰ (Δ = ∼−4‰) at anomaly 1c. Values for Nonionellina labradorica decrease to −8.97‰ (Δ = ∼−7‰) at anomaly 1a, −8.22‰ (Δ = ∼−7‰) at anomaly 1b, and −6.05‰ (Δ = ∼−5‰) at anomaly 1c. Globobulimina auriculata values decrease to −10.79‰ (Δ = ∼−8‰) at anomaly 1a, −12.16‰ (Δ = ∼−10‰) at anomaly 1b, and −7.28‰ (Δ = ∼−5‰) at anomaly 1c. Thus all species analyzed exhibit very negative values during these anomalies although interspecies differences are apparent.


Figure 8. Δ14C and δ13C records of planktonic and benthic foraminifera in δ13C anomaly horizons 1a–1c in the core. (a) Δ14C values of planktonic (N. pachyderma) and benthic (U. akitaensis) foraminifera. Dashed lines represent regression lines of Δ14C of normal planktonic and benthic foraminifera collected before and after horizons of 13C-depleted foraminiferal excursions in relation to depth. (b) The δ13C records of planktonic foraminifera (N. pachyderma, G. bulloides). (c) The δ13C records for benthic foraminifera (U. akitaensis, E. batialis, Nonionellina labradorica, and G. auriculata).

Download figure to PowerPoint


Figure 9. (a) The 14C age differences between anomalous and normal planktonic (N. pachyderma) and benthic (U. akitaensis) foraminifera with depth in the core. (b) The 14C ages of TOC and anomalous and normal planktonic (N. pachyderma) and benthic (U. akitaensis) foraminifera with depth in the core; 1a, 1b, 1c, and 3 represent the anomalous horizons.

Download figure to PowerPoint

The Δ14C values of sedimentary OC were analyzed in horizons that were similar to the anomalous horizons. The TOC ages for marine sediments are generally known to be older than those of planktonic and benthic foraminifera because marine sedimentary organic carbon derives from various carbon sources with different residence time [Eglinton et al., 1997; Uchida et al., 2001, 2005, 2006]. The youngest ages in the same horizons are most likely organic carbon derived from marine autotrophic and heterotrophic organisms in the surface water, which generally provide ages similar to those of planktonic foraminifera using DIC of the surface water [Pearson et al., 2000; Smittenberg et al., 2004]. Sedimentary OC can also be derived from complex and heterogeneous carbon sources from terrestrial reservoirs such as plant-derived lipids, black carbon, and kerogen. Carbon from black carbon and kerogen is almost 14C free, even in the surface horizons of the sediment core [Dickens et al., 2004, 2006]. Carbon from plant-derived lipids are several thousands years older than marine primary organisms [Eglinton et al., 1997; Uchida et al., 2001, 2005]. Our results showing a difference between TOC 14C and that of planktonic foraminifera in the glacial sediments, shown in Figure 9, are thus consistent with this general knowledge. However, TOC 14C values fluctuate greatly around the anomalous horizons. We do not have clear answers to explain these fluctuations and will need to further study the composition of sedimentary organic matter. A molecular biomarker analysis of these sediments is ongoing.

Some researchers have reported that authigenic carbonate may be affected by in situ diagenesis of sedimentary organic matter because sedimentary OC has an extremely 13C-depleted composition (−20∼−25‰) relative to that of foraminifera [Cannariato and Stott, 2004]. In the next section, we investigate relative contribution from sedimentary OC oxidation to foraminiferal anomaly using a mass balance calculation. Bioturbation causes divergent 14C values in foraminifera [Broecker et al., 2006]. This influence of bioturbation appears generally limited to areas of the open ocean, where sedimentation rates are less than 10 cm/ka [Broecker et al., 2006]. At this study site, sedimentation rates across the anomalous horizons were extremely high, ranging from 33 to 187 cm/ka, as determined from normal planktonic foraminifera. These high sedimentation rates in this study site thus are able to likely preclude the influence of bioturbation.

3.7. Coupled 14C and 13C Isotopic Mass Balance

To evaluate the inputs of carbon from methane and seawater DIC during the formation of authigenic carbonate and/or primary biological calcite, we developed an isotopic mass balance model using Δ14C and δ13C. This coupled 14C and 13C isotopic mass balance approach is being used for the first time in regards to the anomalous foraminifera in this study. The approach also provides an adequate screening for methane-induced early diagenesis and is a breakthrough for understanding the drastic warming that occurred during the last glacial episode and was associated with seafloor methane hydrate reservoir instability. Specifically, this model quantitatively tracks contributions to carbonate alkalinity using the signature, end-member isotopic compositions of the various carbon pools relative to bulk isotopic compositions of authigenic carbonate of anomalous foraminifera formed from this alkalinity. In this model, we simplify the carbon sources by dividing them into two possible input pools: one from CaCO3 (authigenic and biogenic) formed from the DIC in seawater and the other representing the DIC generated during the oxidation of methane. The following equations provide the basis for the model, where F refers to the fraction of each of the two carbon sources expressed as the Δ14C and δ13C values of the authigenic CaCO3 of anomalous foraminifera.

  • equation image
  • equation image

From equation (3), the total carbon inputs from the two possible sources must total 1 (i.e., 100%). This expression is then combined with equation (4) to determine the relative contributions of each pool that would result in the measured Δ14C values of the authigenic carbonate of the anomalous foraminifera. The model results are shown in Table 3. In order to obtain the relative contribution (Fmethane) of carbon derived from the oxidation of hydrate methane, we assumed the Δ14C of methane to be Δ14Cmethane = −1000‰. The Δ14Cseawater values for each horizon were calculated from a linear regression of Δ14C values of normal foraminifera against depth from the horizons that span the anomalous horizons. To obtain δ13Cmethane, we used the following isotopic compositions in the model: the δ13Cseawater for planktonic foraminifera and benthic foraminifera were −0.5‰ and −1.5‰, respectively. The mass balance equation for the two carbon pools is similar to equation (4):

  • equation image
Table 3. Modeled Results of Relative Abundance of Methane-Derived Carbon in the Anomalous Foraminiferal Tests and Their δ13C Values Using Coupled 13C and 14C Isotopic Mass Balance Approach Using Δ14Cmethane = −1000‰
Depth in Core, cmAnomaly EventPlanktonic Foraminifera (N. pachyderma)Benthic Foraminifera (U. akitaensis)
Anomaly δ13C, ‰Inferred Δ14Ca (a), ‰Measured Δ14C (b), ‰Δ14C Difference (b)-(a), ‰Fmethane,b wt.%Predicted δ13Cmethane, ‰Anomaly δ13C, ‰Inferred Δ14Ca (a), ‰Measured Δ14C (b), ‰Δ14C Difference (b)-(a), ‰Fmethane,b wt.%Predicted δ13Cmethane, ‰
  • a

    These values represent estimated Δ14C values of normal foraminifera in the same horizons contained anomalous foraminifera, which is calculated from the linear regression line using measured Δ14C data of normal foraminifera.

  • b

    Relative contribution (Fmethane) of methane-derived carbon in total foraminiferal carbonate.


Using coupled 14C and 13C isotopic mass balance techniques, we calculated the relative contribution of carbon from authigenic carbonate and/or primary biogenic calcite that was apparently derived from methane hydrate to be between 8 wt.% and 22 wt.% of total carbonate for planktonic foraminifera and to be between 3 wt.% and 15 wt.% of that for benthic foraminifera. These contributions are consistent with results estimated by acid incremental-leaching tests of authigenic carbonate associated with foraminiferal isotopic anomalies [Millo et al., 2005b]. Thus this model is a useful approach for identifying the carbon source of 13C-depleted foraminifera.

The δ13Cmethane from the isotopic anomaly data were between −24‰ and −68‰ for planktonic foraminifera and between −40‰ and −108‰ for benthic foraminifera. These values are consistent with reported values (ranging from −55‰ to −110‰) of abiogenic and thermogenic methane from marine sediments [Alperin et al., 1988; Blair and Aller, 1995; Blair et al., 1994; Whiticar et al., 1986; Whiticar, 1999]. The estimated δ13C value developed from a model using observed δ13C values of DIC of pore waters from the deep-sea cold seep at the Hatsushima site in Sagami Bay, Japan [Masuzawa et al., 1995] was about −45‰, which is consistent with our results. In our model, we assumed a Δ14C of methane of −1000‰ (14C free) indicating that there is no contribution of a recent (14C-active) organic carbon reservoir to the hydrate carbon pool. Therefore the organic matter from which the methane hydrate originated must have been buried more than ∼50 ka ago. Recent study from radiocarbon measurement of methane hydrate collected in Hydrate Ridges supports our assumption [Winckler et al., 2002]. In vertical observations of the pore water DIC of surface sediments of the cold seep site in Sagami Bay, Japan [Masuzawa et al., 1995] found highly depleted Δ14C values of DIC (e.g., −938‰; corresponding age, ∼45,000 years B.P.), indicating that the oxidized methane was virtually dead and supplied from the deep layer. Highly depleted Δ14C values of close to −1000‰ have also been reported from authigenic carbonate formed in modern cold seeps [Aloisi et al., 2004]. In addition to carbon from methane hydrate and/or dissolved methane, there is the carbon pool of sedimentary organic matter. In this study, we considered this carbon source using the data of Δ14C values of sedimentary OC (Table 1 and Figure 8). The age differences between normal foraminifera and sedimentary OC were large as compared with anomalous foraminifera, but the contribution of sedimentary OC as additional carbon sources to anomalous foraminifera is not as large in this model calculation. If we consider that sedimentary OC had been the dominant carbon source instead of methane, the relative contributions of the sedimentary OC are much higher (from 36 wt% to 80 wt% of total foraminiferal carbonate), and their modeled δ13C values are also much higher (13C composition ranging from −14‰ to −4‰). Since this combination of relative contributions and δ13C values of sedimentary OC appears to be inconsistent with respect to marine organic matter, in our calculation we assumed that the carbon contribution from marine organic matter was negligible. Additionally, as the TOC in our site is very low, ranging from 0.53% to 0.63%, the oxidation of sedimentary OC was most likely not affected to carbonate alkalinity, inducing authigenic carbonate precipitation derived from a carbon source of sedimentary OC.

Our model indicates that highly depleted δ13C of anomalous foraminifera found in this study site suggest an indirect record of enhanced incorporation of 13C-depleted CO2 formed by a methane oxidation process using 12C-enriched methane derived from methane hydrate dissociation as the main source of carbon. Additionally, there was a small difference between the δ13Cmethane values estimated from planktonic and benthic foraminifera even though they are preserved in the same horizon. This difference suggests that different process between planktonic and benthic foraminifera regarding the formation of authigenic carbonate and/or primary biogenic carbonate may exist. For instance, this difference may reflect variations of the carbon isotopic values of DIC produced by various methane oxidation processes such as aerobic and anaerobic oxidation and extent of oxidation occurring before uptake and/or carbonate precipitation in foraminifera. Additionally, there is the effect of methane-influenced pore waters on foraminiferal distribution and carbonate geochemistry. Indeed, this effect induced much greater heterogeneity in δ13C values in the living foraminifera carbonate in modern methane seepages. Similar effects may influence the δ13C values in fossil foraminifera [Rathburn et al., 2000, 2003]. The more depleted δ13C values of methane estimated for benthic foraminifera compared with planktonic foraminifera in model results may reflect large isotopic fractionation by anaerobic oxidation of methane (AOM). The archaeol and hydroxyarchaeol of molecular biomarkers suggesting the AOM process exists were detected in these horizons (M. Uchida et al., manuscript in preparation, 2008; data not shown). As we do not have a clear explanation for this difference from our small data set, further study is needed to constrain the parameters and data points used in this model.

The results of our coupled 14C and 13C mass balance model show, for the first time, that carbon sources of 13C-depleted foraminifera were mainly derived from oxidation of methane hydrate and/or dissolved 14C-free methane in the environment from the sediment-water interface to the water column.

3.8. Archaeal Community Structures

A total of 97 sequences of archaeal 16S rRNA genes from five sediment layers including anomalous foraminifera were analyzed. The results of a similarity analysis showed that the archaeal communities were mainly composed of the Deep-Sea Archaeal Group (DSAG), Marine Benthic Group-D (MBG-D), and Miscellaneous Crenarchaeota Group (MCG) species (Table 4). A few clones were affiliated with the Terrestrial Miscellaneous Euryarchaeota Group (TMEG), Thermoplasmatales, and South African Gold Mine Euryarchaeota Group (SAGMEG) as minor components. DSAG shared 47–70% of clone sequences in libraries from three sediment samples (100–128, 190–212, and 230–247 cm), showing maxima in the layer (230–247 cm) in which 13C-depleted foraminifera were detected, while MBG-D and MCG dominated clone libraries from two deeper layers (276–313 and 520–594 cm, respectively; Table 4).

Table 4. Archaeal Community Structures Based on 16S rRNA Gene Sequences
Representative Sequence100–128 cmbsf190–212 cmbsf230–247 cmbsf276–313 cmbsf520–594 cmbsfNumber of Related Clones (>97%)Relative SequenceAccession NumberSimilarity, %Phylogenetic Groupa
  • a

    DSAG, Deep-Sea Archaeal Group; MBG-D, Marine Benthic Group-D; MCG, Miscellaneous Crenarchaeota Group; TMEG, Terrestrial Miscellaneous Euryarchaeota Group; SAGMEG, South African Gold Mine Euryarchaeota Group.

a5.07  11 2OHKA1.8AB09451898DSAG
a4.19  1  1CRA8-27cmAF11912895DSAG
a5.1453610 24NANK-A83AY43652597MBG-D
a5.09   2 2OHKA1.30AB09452699MBG-D
a5.16   1 1NANK-A149AY43652398MBG-D
a5.2433 1411MA-C1-3AY09345099MCG
a3.14 1  910MA-C1-5AY09345198MCG
a2.021  315OHKA15.43AB09456198MCG
a5.08   1 1Arc.171AF00576589MCG
a6.08    11MA-A1-1AY09344693MCG
a6.10    11MA-C1-5AY09345189MCG
a6.13    11MA-B1-3AY09344797MCG
a3.21 1   1OHKA4.77AB09454291TMEG
a2.151    1AMOS1A_4113_D05AY32322097Thermoplasmatales
a6.21    11OHKA4.4AB09453599SAGMEG
Number of total clones1916202220     

Previous molecular studies of archaea in deep marine subsurface sediments suggested that the members of DSAG are the predominant archaeal components of methane hydrate-bearing marine sediments on the Pacific Ocean margin regardless of the location [Inagaki et al., 2006]. On the basis of an RNA-based molecular survey, DSAG also dominated at the sulfate–methane transition zone at ODP Site 1227 off Peru, whereas MCG dominated clone libraries from other layers [Biddle et al., 2006; Sørensen and Teske, 2006]. Although carbon and energy metabolisms of subseafloor microbial components are largely unknown, it is postulated that DSAG plays a role in the biogeochemical carbon cycle associated with methane and preferentially inhabits sediments above methane hydrates and/or sulfate-methane transition zones. Indeed, we detected DSAG-dominated archaeal communities from a layer in which unusual foraminiferal signals were detected. These results are somewhat consistent with paleoenvironmental conditions; however, we have to note that it is currently uncertain whether these archaeal signals reflect extant cells, deeply buried relicts, or remaining mixed populations [Inagaki et al., 2005].

3.9. Implications Concerning the Carbon Source of the Anomalous Foraminiferal Tests

In this study, we obtained the first radiocarbon data of anomalous foraminifera and a representative suite of geochemical and biogeochemical data to estimate the carbon sources of 13C-depleted foraminiferal tests found in the glacial sediments in the western North Pacific. We found evidence of authigenic carbonate and/or primary biogenic calcite by an incremental-leaching examination. We also found a significant amount of Mg calcite on the surface of the foraminiferal tests, which is used as a useful clue for understanding the formation of authigenic carbonate in a modern cold seep environment [Torres et al., 2003]. Additionally, the 14C ages of 13C-depleted foraminiferal tests were all significantly older than those of normal foraminifera found in the layers spanning the layers in which the 13C-depleted foraminifera were found. These results suggest that methane hydrate derived methane (presumably 14C-free carbon) is a dominant carbon source of the anomalous signal in the carbonate fossils. Model results using 14C and 13C isotopic signatures show that the additional carbon content of anomalous foraminiferal tests reached ∼22 wt%. The range of predicted δ13C values of methane was −24 to −108‰, so it is unknown if this methane was formed thermogenically, biogenically, or abiogenically. Our model, with its small but representative data set, demonstrated that methane-derived carbon is a primary source of carbon in the anomalous foraminifera.

As previously mentioned, BSR is widely distributed in the Tokachi forearc basin (Figures 1b and 2). The strata underlying this study site, as shown on the SRP map, may give useful information regarding events associated with past methane hydrate dissociation. The overlying strata on the anticline are characterized by numerous pipe structures or reflection discontinuities (hands in Figure 2). These numerous pipe structures might be related to the gas seepage, but we do not have any evidence showing cold seep-like environments in this area. Since great plate boundary earthquakes of moment magnitude 8.4 occur in this area about every 500 years [Nanayama et al., 2003], the marine strata in this area have been affected by these episodic rupture events. The methane hydrate layers might be episodically destabilized if catastrophic earthquakes of more than moment magnitude 8.4 occurred during the glacial lower sea level. In this case, a great amount of methane gas could dramatically explode through the numerous pipe structures in the overlying strata upward into the seafloor. Foraminiferal δ13C anomalies similar to those found in our study have been reported from a nearby site within the BSR distribution area [Oba, 2002; Sagawa et al., 2006]. We speculate that methane emission may not occur via small-scale events such as a cold seep, but rather via regional events over the entire BSR area off Hokkaido. Recently a significant amount of fossil methane emission from the seafloor associated with a past earthquake event was reported in the Cariaco Basin, where fossil methane is kept in the water column because of stratification in the Cariaco Basin water [Kessler et al., 2005]. Methane hydrate derived methane emission similar to Cariaco Basin may have occurred in this study site.

Alternatively, temporal methane hydrate dissociation may have occurred as a result of drastic climate changes, including the last deglacial warming. The previously mentioned numerous pipe structures seen in overlying strata may make methane hydrate dissociation progressively easier. The top of the anticline can be interpreted as a depressed structure, and the Miocene strata show scattered reflections that could be related to past gas seepage, but the uppermost reflectors do not show pockmarks or diapirs. Moreover, our finding that DSAG in the Archaeal community was the dominant species in the anomalous layer appears to strongly support our scenario of a past methane hydrate instability event of unknown extent and magnitude. This study gives the first data set allowing quantitative consideration of the carbon sources of anomalous foraminiferal tests using species-specific radiocarbon analysis, but further studies are needed to consider model sensitivity and parameters.

4. Conclusion

  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results and Discussion
  6. 4. Conclusion
  7. Acknowledgments
  8. References

Between 17,840 cal years B.P. and 22,600 cal years B.P., sediments show several spikes of 13C-depleted foraminifera. Using coupled 14C and 13C isotopic mass balance techniques, we calculated the relative contribution of carbon from authigenic carbonate and/or primary biogenic calcite apparently derived from methane hydrate dissociation to be between 8 wt% and 22 wt% of total carbonate for planktonic foraminifera and between 3 wt% and 15 wt% for benthic foraminifera. These contributions are consistent with results estimated by the acid incremental-leaching test. Thus this model shows a useful approach for identifying the carbon source of 13C-depleted foraminifera, which appears to be derived from the oxidation of methane hydrate and/or dissolved methane. The δ13Cmethane for planktonic and benthic foraminifera were between −24‰ and −68‰ and −40‰ and −108‰, respectively.

Our model indicates that the highly depleted δ13C results suggest indirect records of enhanced incorporation of 13C-depleted CO2 formed by a methane oxidation process that uses 12C-enriched methane derived from methane hydrate dissociation as a main source of carbon. Moreover, our comparison of modeled δ13C values of planktonic and benthic foraminifera preserved in the same horizon indicates that different processes for the formation of authigenic carbonate and/or primary biogenic carbonate may have existed. The results of a coupled 14C and 13C mass balance model show that the carbon source of 13C-depleted foraminifera was partially derived from methane hydrate and/or dissolved 14C-free methane in the environment from the sediment-water interface to the top of the water column.

In recent paleoceanographic studies, high-sedimentation-rate cores from marginal seas have been successfully used for the high time resolution reconstruction of past climate change [e.g., Kennett et al., 2000]. For this purpose, foraminiferal isotopic records are crucial for age model construction with radiocarbon and stable oxygen isotopes. Thus we should know the influence of authigenic carbonate with respect to the foraminiferal tests used for the age models. The influence of authigenic carbonate is especially apparent in the marginal sea where methane hydrate is widely distributed and any event associated with hydrate dissociation may greatly influence foraminiferal carbonate chemistry. Clearly, a correct depositional age model is necessary for a true reconstruction of past climate change from sediment cores. This study may provide important information for interpreting geological records including methane hydrate instability associated with past climate change and for constructing a correct depositional age model for high-resolution paleoceanographic studies.


  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results and Discussion
  6. 4. Conclusion
  7. Acknowledgments
  8. References

We greatly thank J. Kennett (University of California, Santa Barbara) for suggestions on interpreting the data. We acknowledge N. Ahagon (Hokkaido University) for interpretation of stable carbon isotopic data and I. Motoyama (University of Tsukuba) for his kind cooperation for this cruise. We thank the captain, crew, Marine Work Japan staff, and scientific party of the MR01-K03 cruise of the R/V Mirai for their cooperation at sea. We are also grateful to T. Kobayashi (NIES-TERRA) for AMS operation, N. Kisen and J. Yoshino (Marine Works Japan, Co. Ltd.) for stable isotope analysis, Y. Nakamura (JAMSTEC) for foraminifera picking, and A. Matsuda (NIES) and M. Suzuki (NIES) for sample preparation of AMS analysis. We also thank T. Chiba (Marine Works Japan, Co. Ltd.) for maintaining a mass spectrometer (MIO, JAMSTEC). We thank T. Okano and A. Taira (CDEX, JAMSTEC) for providing detailed BSR investigation data for this study area. Seismic data were obtained during the research cruise GH02 aboard the R/V Hakurei Maru No. 2, within the framework of the Japanese national program “Marine Geological Study of Continental Shelves of the Collisional Area Between Hokkaido and the Kurile Arc,” which is supported by the Geological Survey of Japan (AIST). We are also thankful for comments by L. D. Labeyrie, J. Lynch-Stieglitz, and anonymous reviewers. This study is part of the “Study on the past marine environmental changes,” sponsored by the Japan Agency for Marine-Earth Science and Technology (JAMSTEC). Financial support for this study was provided by the Japanese Ministry of Education, Culture, Sports, Science and Technology, Grant-in Aid for Scientific Research (KAKENHI), Kiban-B-18310019 and Houga-18651010.


  1. Top of page
  2. Abstract
  3. 1. Introduction
  4. 2. Material and Methods
  5. 3. Results and Discussion
  6. 4. Conclusion
  7. Acknowledgments
  8. References
  • Aloisi, G., K. Wallmann, R. R. Haese, and J. F. Saliege (2004), Chemical, biological and hydrological controls on the C-14 content of cold seep carbonate crusts: Numerical modeling and implications for convection at cold seeps, Chem. Geol., 213(4), 359383.
  • Alperin, M. J., W. S. Reeburgh, and M. J. Whiticar (1988), Carbon and hydrogen isotope fractionation resulting from anaerobic methane oxidation, Global Biogeochem. Cycles, 2(3), 279288.
    Direct Link:
  • Bard, E. (1998), Geochemical and geophysical implications of the radiocarbon calibration, Geochim. Cosmochim. Acta, 62, 20252038.
  • Biddle, J. F., et al. (2006), Heterotrophic Archaea dominate sedimentary subsurface ecosystems off Peru, Proc. Natl. Acad. Sci. U. S. A., 103, 38463851.
  • Blair, N. E., and R. C. Aller (1995), Anaerobic methane oxidation on the Amazon Shelf, Geochim. Cosmochim. Acta, 59(18), 37073715.
  • Blair, N. E., G. R. Plaia, S. E. Boehme, D. J. Demaster, and L. A. Levin (1994), The remineralization of organic carbon on the North Carolina continental slope, Deep Sea Res., Part II, 41(4–6), 755766.
  • Boyle, E. A., and L. D. Keigwin (1985), Comparison of Atlantic and Pacific paleochemical records for the last 215,000 years—Changes in deep ocean circulation and chemical inventories, Earth Planet. Sci. Lett., 76(1–2), 135150.
  • Broecker, W., S. Barker, E. Clark, I. Hajdas, and G. Bonani (2006), Anomalous radiocarbon ages for foraminifera shells, Paleoceanography, 21, PA2008, doi:10.1029/2005PA001212.
  • Cannariato, K. G., and L. D. Stott (2004), Evidence against clathrate-derived methane release to Santa Barbara Basin surface waters? Geochem. Geophys. Geosyst., 5, Q05007, doi:10.1029/2003GC000600.
  • Center for Deep Earth Exploration (CDEX) (2002), CDEX site survey data, Jpn. Agency for Mar. Earth Sci. and Technol., Yokosuka, Japan.
  • de Garidel-Thoron, T., L. Beaufort, F. Bassinot, and P. Henry (2004), Evidence for large methane releases to the atmosphere from deep-sea gas-hydrate dissociation during the last glacial episode, Proc. Natl. Acad. Sci. U. S. A., 101(25), 91879192.
  • DeLong, E. F. (1992), Archaea in coastal marine environments, Proc. Natl. Acad. Sci. U. S. A., 89, 56855689.
  • Dickens, A. F., Y. Gelinas, C. A. Masiello, S. Wakeham, and J. I. Hedges (2004), Reburial of fossil organic carbon in marine sediments, Nature, 427(6972), 336339.
  • Dickens, A. F., J. A. Baldock, R. J. Smernik, S. G. Wakeham, T. S. Arnarson, Y. Gelinas, and J. I. Hedges (2006), Solid-state C-13 NMR analysis of size and density fractions of marine sediments: Insight into organic carbon sources and preservation mechanisms, Geochim. Cosmochim. Acta, 70(3), 666686.
  • Dickens, G. R., J. R. O'Neil, D. K. Rea, and R. M. Owen (1995), Dissociation of oceanic methane hydrate as a cause of the carbon isotope excursion at the end of the Paleocene, Paleoceanography, 10, 965971.
  • Dickens, G. R., M. M. Castillo, and J. C. Walker (1997), A blast of gas in the latest Paleocene: Simulating first-order effects of massive dissociation of oceanic methane hydrate, Geology, 25(3), 259262.
  • Duplessy, J.-C., M. Arnold, E. Bard, A. Juillet-Leclerc, N. Kallel, and L. Labeyrie (1989), AMS 14C study of transient events and of the ventilation rate of the Pacific intermediate water during the last deglaciation, Radiocarbon, 31, 493502.
  • Eglinton, T. I., C. Bryan, A. Benitez-Nelson, A. P. Pearson, J. E. McNichol, Z. Bauer, and R. M. Druffel (1997), Variability in radiocarbon ages of individual organic compounds from marine sediments, Science, 277, 796799.
  • Elderfield, H., and G. Ganssen (2000), Past temperature and δ18O of surface ocean waters inferred from foraminiferal Mg/Ca ratios, Nature, 405(6785), 442445.
  • Formolo, M. J., T. W. Lyons, C. L. Zhang, C. Kelley, R. Sassen, J. Horita, and D. R. Cole (2004), Quantifying carbon sources in the formation of authigenic carbonates at gas hydrate sites in the Gulf of Mexico, Chem. Geol., 205(3–4), 253264.
  • Grabowski, K. S., D. L. Knies, S. J. Tumey, J. W. Pohlman, C. S. Mitchell, and R. B. Coffin (2004), Carbon pool analysis of methane hydrate regions in the seafloor by accelerator mass spectrometry, Nucl. Instrum. Methods Phys. Res., Sect. B, 223–24, 435440.
  • Hasselbo, S. P., D. R. Grocke, H. C. Jenkyns, C. J. Bjerrum, P. Farrimond, H. S. M. Bell, and O. R. Green (2000), Massive dissociation of gas hydrate during a Jurassic oceanic anoxic event, Nature, 406, 392395.
  • Hill, T. M., J. P. Kennett, and H. J. Spero (2004a), High-resolution records of methane hydrate dissociation: ODP Site 893, Santa Barbara Basin, Earth Planet. Sci. Lett., 223(1–2), 127140.
  • Hill, T. M., J. P. Kennett, and D. L. Valentine (2004b), Isotopic evidence for the incorporation of methane-derived carbon into foraminifera from modern methane seeps, Hydrate Ridge, northeast Pacific, Geochim. Cosmochim. Acta, 68(22), 46194627.
  • Hill, T. M., J. P. Kennett, D. L. Valentine, Z. Yang, C. M. Reddy, R. K. Nelson, R. J. Behl, C. Robert, and L. Beaufort (2006), Climatically driven emissions of hydrocarbons from marine sediments during deglaciation, Proc. Natl. Acad. Sci. U. S. A., 103(37), 13,57013,574.
  • Hinrichs, K. U., L. R. Hmelo, and S. P. Sylva (2003), Molecular fossil record of elevated methane levels in late Pleistocene coastal waters, Science, 299(5610), 12141217.
  • Hoshiba, M., N. Ahagon, K. Ohkushi, M. Uchida, I. Motoyama, and A. Nishimura (2006), Foraminiferal oxygen and carbon isotopes off north Japan, northwestern Pacific during the last 34 kyr, Mar. Micropaleontol., 61, 196208.
  • Inagaki, F., M. Suzuki, K. Takai, H. Oida, T. Sakamoto, K. Aoki, K. H. Nealson, and K. Horikoshi (2003), Microbial communities associated with geological horizons in coastal subseafloor sediments from the Sea of Okhotsk, Appl. Environ. Microbiol., 69, 72247235.
  • Inagaki, F., H. Okada, A. I. Tsapin, and K. H. Nealson (2005), The Paleome: A genetic record of bacterial communities in mid-Cretaceous black shale, Astrobiology, 5, 141153.
  • Inagaki, F., et al. (2006), Biogeographical distribution and diversity of microbes in methane hydrate-bearing deep marine sediments, on the Pacific Ocean Margin, Proc. Natl. Acad. Sci. U. S. A., 103(8), 28152820.
  • Ishimura, T., U. Tsunogai, and T. Gamo (2004), Stable carbon and oxygen isotopic determination of sub-microgram quantities of CaCO3 to analyze individual foraminiferal shells, Rapid Commun. Mass Spectrom., 18, 28832888.
  • Jahren, A. H., N. C. Arens, G. Sarmiento, J. Guerrero, and R. Amundson (2001), Terrestrial record of methane hydrate dissociation in the Early Cretaceous, Geology, 29(2), 159162.
  • Keigwin, L. D. (2002), Late Pleistocene-Holocene paleoceanography and ventilation of the Gulf of California, J. Oceanogr., 58, 421432.
  • Kennett, J. P., K. G. Cannariato, I. L. Hendy, and R. J. Behl (2000), Carbon isotopic evidence for methane hydrate instability during Quaternary interstadials, Science, 288(5463), 128133.
  • Kessler, J. D., W. S. Reeburgh, J. Southon, and R. Varela (2005), Fossil methane source dominates Cariaco Basin water column methane geochemistry, Geophys. Res. Lett., 32, L12609, doi:10.1029/2005GL022984.
  • Kessler, J. D., W. S. Reeburgh, and S. C. Tyler (2006), Controls on methane concentration and stable isotope (δ2H-CH4 and δ13C-CH4) distributions in the water columns of the Black Sea and Cariaco Basin, Global Biogeochem. Cycles, 20, GB4004, doi:10.1029/2005GB002571.
  • Kvenvolden, K. A., and T. D. Lorenson (2001), The global occurrence of natural gas hydrates, in Natural Gas Hydrates: Occurrence, Distribution, and Detection, Geophys. Monogr. Ser., vol. 124, edited by C. K. Paull, and W. P. Dillon, pp. 318, AGU, Washington, D. C.
  • Lea, D. W., T. A. Mashiotta, and H. J. Spero (1999), Controls on magnesium and strontium uptake in planktonic foraminifera determined by live culturing, Geochim. Cosmochim. Acta, 63(16), 23692379.
  • Maslin, M., N. Mikkelsen, C. Vilela, and B. Haq (1998), Sea-level- and gas-hydrate-controlled catastrophic sediment failures of the Amazon Fan, Geology, 26(12), 11071110.
  • Masuzawa, T., H. Kitagawa, T. Nakatsuka, N. Handa, and T. Nakamura (1995), AMS C-14 measurements of dissolved inorganic carbon in pore waters from a deep-sea “cold seep” giant clam community off Hatsushima Island, Sagami Bay, Japan, Radiocarbon, 37(2), 617627.
  • McNichol, A. P., E. A. Osborne, A. R. Gagnon, B. Fry, and G. A. Jones (1994), Tic, Toc, Dic, Doc, Pic, Poc—Unique aspects in the preparation of oceanographic samples for C-14 AMS, Nucl. Instrum. Methods Phys. Res., Sect. B, 92(1–4), 162165.
  • Millo, C., M. Sarnthein, H. Erlenkeuser, and H. Frederichs (2005a), Methane-driven late Pleistocene δ13C minima and overflow reversals in the southwestern Greenland Sea, Geology, 33(11), 873876.
  • Millo, C., M. Sarnthein, H. Erlenkeuser, P. M. Grootes, and N. Andersen (2005b), Methane-induced early diagenesis of foraminiferal tests in the southwestern Greenland Sea, Mar. Micropaleontol., 58(1), 112.
  • Nagao, S., T. Usui, M. Yamamoto, M. Minagawa, T. Iwatsuki, and A. Noda (2005), Combined use of delta C-14 and delta C-13 values to trace transportation and deposition processes of terrestrial particulate organic matter in coastal marine environments, Chem. Geol., 218(1–2), 6372.
  • Nanayama, F., K. Satake, R. Furukawa, K. Shimokawa, B. F. Atwater, K. Shigeno, and S. Yamaki (2003), Unusually large earthquakes inferred from tsunami deposits along the Kuril trench, Nature, 424(6949), 660663.
  • Nurnberg, D., J. Bijma, and C. Hemleben (1996), Assessing the reliability of magnesium in foraminiferal calcite as a proxy for water mass temperatures, Geochim. Cosmochim. Acta, 60(5), 803814.
  • Oba, T. (2002), Paleoceanographic changes off the east coast of the Japanese islands during the last 20 ka inferred from oxygen and carbon isotope of foraminiferal tests (in Japanese), paper presented at 32nd Meeting of the Japan Association for Quaternary Research, Matsumoto, Japan.
  • Ohkushi, K., N. Ahagon, M. Uchida, and Y. Shibata (2005), Foraminiferal isotope anomalies from northwestern Pacific marginal sediments, Geochem. Geophys. Geosyst., 6, Q04005, doi:10.1029/2004GC000787.
  • Pearson, A., T. I. Eglinton, and A. P. McNichol (2000), An organic tracer for surface ocean radiocarbon, Paleoceanography, 15(5), 541550.
  • Prokopenko, A. A., and D. F. Williams (2004), Deglacial methane emission signals in the carbon isotopic record of Lake Baikal, Earth Planet. Sci. Lett., 218(1–2), 135147.
  • Prokopenko, A. A., and D. F. Williams (2005), Depleted methane-derived carbon in waters of Lake Baikal, Siberia, Hydrobiologia, 544, 279288.
  • Rathburn, A. E., L. A. Levin, Z. Held, and K. C. Lohmann (2000), Benthic foraminifera associated with cold methane seeps on the Northern California margin: Ecology and stable isotopic composition, Mar. Micropaleontol., 38, 247266.
  • Rathburn, A. E., M. E. Pérez, J. B. Martin, S. A. Day, C. Mahn, J. Gieskes, W. Ziebis, D. Williams, and A. Bahls (2003), Relationships between the distribution and stable isotopic composition of living benthic foraminifera and cold methane seep biogeochemistry in Monterey Bay, California, Geochem. Geophys. Geosyst., 4(12), 1106, doi:10.1029/2003GC000595.
  • Rosenthal, Y., M. P. Field, and R. M. Sherrell (1999), Precise determination of element/calcium ratios in calcareous samples using sector field inductively coupled plasma mass spectrometry, Anal. Chem., 71(15), 32483253.
  • Sagawa, T., M. Murayama, T. Oba, and K. Ikehara (2006), Millennial scale displacement of the subarctic boundary during the last deglaciation in the northwestern Pacific, Eos Trans. AGU, 87(52), Fall Meet. Suppl., Abstract PP13B-1602.
  • Sasaki, A., T. Kachi, T. Sasaoka, and T. Iguchi (1985), Stratigraphy of the Kisoshisui Tokachi-oki well on the study of Miocene turbidite facies in eastern Hokkaido (in Japanese with English abstract), J. Jpn. Assoc. Petrol. Technol., 50, 5363.
  • Satoh, M. (1994), Geophysical distribution of naturally occurring gas hydrates, Earth Mon., 16, 533538.
  • Satoh, M. (2002), Distribution and resources of marine natural gas hydrates around Japan, paper presented at Fourth International Conference on Gas Hydrates, Mitsui Eng. and Shipbuilding Co., Ltd., Yokohama, Japan.
  • Schimmelmann, A. (1991), Determination of the concentration and stable isotopic composition of nonexchangeable hydrogen in organic matter, Anal. Chem., 63(21), 24562459.
  • Smith, L. M., J. P. Sachs, A. E. Jennings, D. M. Anderson, and A. deVernal (2001), Light δ13C events during deglaciation of the east Greenland continental shelf attributed to methane release from gas hydrates, Geophys. Res. Lett., 28(11), 22172220.
  • Smittenberg, R. H., E. C. Hopmans, S. Schouten, J. M. Hayes, T. I. Eglinton, and J. S. Sinninghe Damsté (2004), Compound-specific radiocarbon dating of the varved Holocene sedimentary record of Saanich Inlet, Canada, Paleoceanography, 19, PA2012, doi:10.1029/2003PA000927.
  • Sørensen, K. B., and A. Teske (2006), Stratified communities of active archaea in deep marine subsurface sediments, Appl. Environ. Microbiol., 72, 45964603.
  • Stuiver, M., P. J. Reimer, E. Bard, J. W. Beck, G. S. Burr, K. A. Hughen, B. Kromer, G. McCormac, J. van der Plicht, and M. Spurk (1998), Intcal98 radiocarbon age calibration, 24,000–0 cal BP, Radiocarbon, 40, 10411083.
  • Taira, A., and D. Curewitz (2005), Shimokita area site survey: Northern Japan Trench seismic survey, offshore northern Honshu, Japan, CDEX Tech. Rep. 2, Cent. for Deep Earth Explor., Jpn. Agency for Mar. Earth Sci. and Technol., Yokosuka, Japan. (Available at
  • Tanaka, A., M. Yoneda, M. Uchida, T. Uehiro, Y. Shibata, and M. Morita (2000), Recent advances in C-14 measurement at NIES-TERRA, Nucl. Instrum. Methods Phys. Res, Sect. B, 172, 107111.
  • Torres, M. E., A. C. Mix, K. Kinports, B. Haley, G. P. Klinkhammer, J. McManus, and M. A. de Angelis (2003), Is methane venting at the seafloor recorded by δ13C of benthic foraminifera shells? Paleoceanography, 18(3), 1062, doi:10.1029/2002PA000824.
  • Torres, M. E., A. C. Mix, and W. D. Rugh (2005), Precise δ13C analysis of dissolved inorganic carbon in natural waters using automated headspace sampling and continuous-flow mass spectrometry, Limnol. Oceanogr. Methods, 3, 349360.
  • TuZino, T., and A. Noda (2007), Tectonic control over topography and channel sedimentation across the forearc slope of the southern Kurile Trench, Geo Mar. Lett., 27(1), 111.
  • Uchida, M., et al. (2001), Compound-specific radiocarbon ages of fatty acids in marine sediments from the western North Pacific, Radiocarbon, 43(2B), 949956.
  • Uchida, M., Y. Shibata, K. Ohkushi, N. Ahagon, and M. Hoshiba (2004), Episodic methane release events from Last Glacial marginal sediments in the western North Pacific, Geochem. Geophys. Geosyst., 5, Q08005, doi:10.1029/2004GC000699.
  • Uchida, M., Y. Shibata, K. Ohkushi, M. Yoneda, K. Kawamura, and M. Morita (2005), Age discrepancy between molecular biomarkers and calcareous foraminifera isolated from the same horizons of Northwest Pacific sediments, Chem. Geol., 218(1–2), 7389.
  • Uchida, M., T. I. Eglinton, L. Coppola, Ö. Gustafsson, P. Andersson, D. Montlucon, and J. M. Hayes (2006), Hydrodynamic controls on the age and composition of terrestrial organic matter distributed over the Washington Margin, Eos Trans. AGU, 87(52), Fall Meet. Suppl., Abstract OS23A-1625.
  • Wefer, G., P. M. Heinze, and W. H. Berger (1994), Clues to ancient methane release, Nature, 369(6478), 282.
  • Whiticar, M. J. (1999), Carbon and hydrogen isotope systematics of bacterial formation and oxidation of methane, Chem. Geol., 161(1–3), 291314.
  • Whiticar, M. J., E. Faber, and M. Schoell (1986), Biogenic methane formation in marine and fresh-water environments—CO2 reduction vs acetate fermentation isotope evidence, Geochim. Cosmochim. Acta, 50(5), 693709.
  • Winckler, G., W. Aeschbach-Hertig, J. Holocher, R. Kipfer, I. Levin, C. Poss, G. Rehder, P. Schlosser, and E. Suess (2002), Noble gases and radiocarbon in natural gas hydrates, Geophys. Res. Lett., 29(10), 1423, doi:10.1029/2001GL014013.
  • Yoneda, M., et al. (2004), AMS C-14 measurement and preparative techniques at NIES-TERRA, Nucl. Instrum. Methods Phys. Res., Sect. B, 223–224, 116123.