Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the...

16
Research Article Linkage of deep sea rapid acidification process and extinction of benthic foraminifera in the deep sea at the Paleocene/Eocene transition HODAKA KAWAHATA, 1, *RITSUO NOMURA, 2 KATSUMI MATSUMOTO, 3 AND HIROSHI NISHI 4 1 Atmosphere and Ocean Research Institute, The University of Tokyo, Kashiwanoha 5-1-5, Kashiwa, Chiba 277-8564, Japan (email: [email protected]), 2 Faculty of Education, Shimane University, 1060 Nishi-kawatsucho, Matsue 690-8504, Japan, 3 Department of Geology and Geophysics, University of Minnesota, 310 Pillsbury Dr. SE, Minneapolis, MN 55455, USA, and 4 The Center for Academic Resources and Archives, Tohoku University Museum, Tohoku University, Aramaki Aza Aoba 6-3, Sendai, 980-8578, Japan Abstract Ocean Drilling Program Leg 199 Site 1220 provides a continuous sedimentary section across the Paleocene/Eocene (P/E) transition in the carbonate-bearing sediments on 56–57 Ma oceanic crust. The large negative δ 13 C shift in seawater is likely due to the disintegration of methane hydrate, which is expected to be rapidly changed to carbon dioxide in the atmosphere and well-oxygenated seawater, leading to a reduction in deep-sea pH. A pH decrease was very likely responsible for the emergence of agglutinated fora- miniferal fauna as calcareous fauna was eliminated by acidification at the P/E transition at Site 1220. The absence of the more resistant calcareous benthic foraminifera and the presence of the planktonic foraminifera at Site 1220 is interesting and unique, which indicates that calcareous benthic foraminifera suffered greatly from living on the seafloor. Box model calculation demonstrates that, assuming the same mean alkalinity as today, pCO 2 must increase from 280 ppm to about 410 ppm for the calcite undersaturation in the deep ocean and for the oversaturation in the surface ocean during the P/E transition. The calculated increased pCO 2 coincides with paleo-botanical evidence. The current global emission rate (7.3 peta (10 15 ) gC/y) of anthropogenic carbon input is approximately 30 times of the estimate at the P/E transition. The results at the P/E transition give an implication that the deep sea benthic fauna will be threatened in future in combination with ocean acidification, increased sea surface temperature and more stratified surface water. Key words: agglutinated foraminifera, benthic foraminifera, carbon cycle, ocean acidifica- tion, Ocean Drilling Program (ODP), The Paleocene/Eocene (P/E) transition. INTRODUCTION The Paleocene/Eocene (P/E) transition, often referred to as the Paleocene–Eocene thermal maximum, is characterized by large excursions or ‘events’ in the oxygen and carbon isotope records (e.g., Kennett & Stott 1991). The duration of these events are estimated to be no more than 10 ky with P/E transition conditions with a total duration of about 215 ± 5 ky for carbon isotope excursion (Röhl et al. 2000). Benthic foraminiferal δ 18 O values indicate that oceanic deep waters warmed substantially worldwide, from about 10 to 18°C in high latitudes (Kennett & Stott 1991) and from about 10 to 16°C in the tropics (Bralower et al. 1995). Sea-surface temperatures (SSTs), as recorded in planktonic foraminiferal δ 18 O and/or Mg/Ca values showed a little change from about 26 to 28°C at low latitudes but larger increase at high latitudes (Kennett & Stott 1991; Tripati & Elderfield 2004), leading to diminished vertical and latitudinal thermal gradients (Kelly et al. 1996). These estimates have been reconfirmed by recent detailed analysis of TEX86 and foraminiferal δ 18 O (Sluijs et al. 2006; John et al. 2008). Actually the *Correspondence. Received 11 August 2014; accepted for publication 3 February 2015. Island Arc (2015) © 2015 Wiley Publishing Asia Pty Ltd doi:10.1111/iar.12106

Transcript of Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the...

Page 1: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

Research ArticleLinkage of deep sea rapid acidification process and extinction of benthic

foraminifera in the deep sea at the Paleocene/Eocene transition

HODAKA KAWAHATA,1,* RITSUO NOMURA,2 KATSUMI MATSUMOTO,3 AND HIROSHI NISHI4

1Atmosphere and Ocean Research Institute, The University of Tokyo, Kashiwanoha 5-1-5, Kashiwa, Chiba277-8564, Japan (email: [email protected]), 2Faculty of Education, Shimane University, 1060

Nishi-kawatsucho, Matsue 690-8504, Japan, 3Department of Geology and Geophysics, University of Minnesota,310 Pillsbury Dr. SE, Minneapolis, MN 55455, USA, and 4The Center for Academic Resources and Archives,

Tohoku University Museum, Tohoku University, Aramaki Aza Aoba 6-3, Sendai, 980-8578, Japan

Abstract Ocean Drilling Program Leg 199 Site 1220 provides a continuous sedimentarysection across the Paleocene/Eocene (P/E) transition in the carbonate-bearing sedimentson 56–57 Ma oceanic crust. The large negative δ13C shift in seawater is likely due to thedisintegration of methane hydrate, which is expected to be rapidly changed to carbondioxide in the atmosphere and well-oxygenated seawater, leading to a reduction in deep-seapH. A pH decrease was very likely responsible for the emergence of agglutinated fora-miniferal fauna as calcareous fauna was eliminated by acidification at the P/E transition atSite 1220. The absence of the more resistant calcareous benthic foraminifera and thepresence of the planktonic foraminifera at Site 1220 is interesting and unique, whichindicates that calcareous benthic foraminifera suffered greatly from living on the seafloor.Box model calculation demonstrates that, assuming the same mean alkalinity as today,pCO2 must increase from 280 ppm to about 410 ppm for the calcite undersaturation in thedeep ocean and for the oversaturation in the surface ocean during the P/E transition. Thecalculated increased pCO2 coincides with paleo-botanical evidence. The current globalemission rate (∼7.3 peta (1015) gC/y) of anthropogenic carbon input is approximately 30times of the estimate at the P/E transition. The results at the P/E transition give animplication that the deep sea benthic fauna will be threatened in future in combination withocean acidification, increased sea surface temperature and more stratified surface water.

Key words: agglutinated foraminifera, benthic foraminifera, carbon cycle, ocean acidifica-tion, Ocean Drilling Program (ODP), The Paleocene/Eocene (P/E) transition.

INTRODUCTION

The Paleocene/Eocene (P/E) transition, oftenreferred to as the Paleocene–Eocene thermalmaximum, is characterized by large excursions or‘events’ in the oxygen and carbon isotope records(e.g., Kennett & Stott 1991). The duration of theseevents are estimated to be no more than 10 ky withP/E transition conditions with a total duration ofabout 215 ± 5 ky for carbon isotope excursion(Röhl et al. 2000). Benthic foraminiferal δ18O

values indicate that oceanic deep waters warmedsubstantially worldwide, from about 10 to 18°C inhigh latitudes (Kennett & Stott 1991) and fromabout 10 to 16°C in the tropics (Bralower et al.1995). Sea-surface temperatures (SSTs), asrecorded in planktonic foraminiferal δ 18O and/orMg/Ca values showed a little change from about 26to 28°C at low latitudes but larger increase at highlatitudes (Kennett & Stott 1991; Tripati &Elderfield 2004), leading to diminished vertical andlatitudinal thermal gradients (Kelly et al. 1996).These estimates have been reconfirmed by recentdetailed analysis of TEX86 and foraminiferal δ 18O(Sluijs et al. 2006; John et al. 2008). Actually the

*Correspondence.

Received 11 August 2014; accepted for publication 3 February 2015.

Island Arc (2015)

© 2015 Wiley Publishing Asia Pty Ltd doi:10.1111/iar.12106

Page 2: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

declines of coralgal reefs could be attributed toSST in the tropics beyond the maximum tempera-ture range of corals (Scheibner & Speijer 2008). Inthe terrestrial environments, a widespreadincrease in kaolinite was interpreted to herald awarm, wet climate (Kennett and Stott, 1991). Inassociation with these environmental changes,ocean circulation, primary productivity, carbonatedissolution and hydrologic cycle showed severedisturbed wide range of environments (Bralower2002; Ernst et al. 2006; Stassen et al. 2012).

Coincident with the P/E transition, the globalcarbon cycle experienced a concomitant huge per-turbation (Zachos et al. 2001). A large δ 13Canomaly, which has been observed in deep-seasediments, terrestrial soil carbonates and mammalteeth associated with a major turnover in landmammal assemblages, was −3‰ (up to −6‰) overan interval of about 100–200 ky near the end of thePaleocene (Kennett & Stott 1991; Koch et al. 1992;Bralower et al. 1997; Pak et al. 1997; Norris & Röhl1999; Pagani et al. 2006). Several explanations of alarge negative shift in δ 13C have been proposed: (i)dissociation of submarine methane hydrates(Dickens et al. 1995; Xu et al. 2001); (ii) volcanismincluding hydrothermal activity on the seafloor(Bralower et al. 1997; Svensen et al. 2004); and (iii)the collision of a comet (Kent et al. 2003). Whilecurrently available data strongly support explana-tions that involve injection of large amounts oforganically derived methane hydrates, the exacttriggering mechanism such as slope disturbanceand/or landslide for the hydrate collapse has notyet been agreed (Katz et al. 1999; Storey et al.2007). The amount of carbon required to explainthe negative δ 13C shift is estimated to be as largeas 2000 peta (1015) gC (Dickens et al. 1997; Zachoset al. 2005).

In the oceans, the δ 13C anomaly coincides withthe most dramatic extinction of 35–50% of cosmo-politan benthic foraminifera in the past 100 millionyears (Tjalsma & Lohmann 1983; Kelly et al. 1996;Thomas & Shackleton 1996) and a rapid shoaling ofthe global CCD (Calcite Compensation Depth)(Zachos et al. 2001; Kroon et al. 2004). However,the degree of extinction(s)/faunal turnovers wasapparently much less severe for continental shelfbenthic foraminifera and minimal for planktonicecosystems, including planktonic foraminiferafaunas, calcareous nanoplankton, and dinoflagel-late floras (Raffi et al. 2005). The specific mecha-nism behind the deep sea benthic foraminiferextinction remains unclear. However, several ideashave been proposed: (i) the formation of oxygen-

deficient deep waters (Kelly et al. 1996; Katz et al.1999); (ii) warm saline bottom water from the lowlatitude (Bice & Marotzke 2001); and (iii) dissolu-tion of deep-sea carbonate (Zachos et al. 2005).

There have been few P/E sections available withaccurate age control because the interval is oftenaffected appreciably by intense carbonate dissolu-tion. However, Ocean Drilling Program (ODP) Leg199 provides a continuous sedimentary sectionacross the P/E boundary in the carbonate-bearingsediments on 56–57 Ma oceanic crust. In thispaper, we present high time resolution recordsfrom Leg 199 of benthic foraminiferal assemblageand sediment geochemistry in order to discuss theplausible causal relationship between deep seabenthic foraminiferal mass extinction and acidifi-cation at the P/E transition with implications forthe future as the global ocean becomes increas-ingly acidic under continued emissions of anthro-pogenic carbon into the atmosphere.

MATERIAL AND METHODS

GEOLOGIC SETTING AND STUDIED CORES

The ODP Leg 199 Site 1220 is currently located ata water depth of 5218 m in typical abyssal hilltopography at 10°10.601′N, 142°45.491′W in theNorth Pacific Ocean (Fig. 1). Based upon a fixedhotspot model (Gripp & Gordon 1990, for the 0- to5-Ma Pacific hotspot rotation pole; Engebretsonet al. 1985, for older poles), Site 1220 must havebeen located at around 3°S of the equator at 56 Ma.Thus Site 1220 must have been affected by theSouth Equatorial Current in the early Eocene andthe late Paleocene. Hole 1220B was drilled intosediments, reaching basement at around 200meters below sea-floor (mbsf). The lower part ofthe sedimentary sequence corresponds to a lowerEocene-upper Paleocene unit (∼185–200 mbsf) ofpartially dolomitized nanofossil ooze, radiolariannanofossil ooze, radiolarian ooze, calcareous chalk,and black clay atop an aphanitic to fine-grainedphaneritic basalt.

TRANSITION AT SITE 1220

Both planktonic foraminifers and calcareousnanofossils provide a detailed zonation of a con-densed sequence of lower Eocene and upper Paleo-cene nanofossil oozes (Shipboard Scientific Partyof Leg 199, 2002). The extinction of Paleocenebenthic foraminifers, the appearance of the

2 H. Kawahata et al.

© 2015 Wiley Publishing Asia Pty Ltd

Page 3: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

nanofossil genus Rhomboaster, the extinction ofthe nanofossil genus Fasciculithus, and the pres-ence of excursion fauna of planktonic foraminiferspresent a detailed biostratigraphy of the P/E tran-sition in Unit 5 (187.9–200.0 mbsf) at Site 1220. Theonset of the carbon isotope excursion, which marksthe Paleocene/Eocene transition, coincides withthe beginning of a benthic foraminiferal gradualextinction in Unit 5 by the comparison betweenSites 1220 and 1221 (Nunes & Norris 2005).

Alternating intervals of white and pale browncalcareous chalk occur between 198.90 and 199.38mbsf and contain a minor amount (5–10%) ofpoorly preserved nanofossils. White layers occurbetween 199.15 and 199.23 mbsf and contain15–30% planktonic foraminifers. The sedimentsbelow this interval consist of faintly banded, darkyellowish brown calcareous chalk devoid of micro-fossils. Dolomite (up to 10%) and clay (10–35%)also occur in this lithology. Below 199.38 mbsf is a12 cm thick layer of very dark brown calcareouschalk underlain by black clay (199.50–199.55).Major components of both sediments are dolo-mite, iron manganese oxides, and clay. Dolomitecontent decreases over the interval from 199.52 to199.54 mbsf, and very limited calcareous or sili-ceous fossils were observed in correspondingsmear slides. Directly beneath the black clay is a4 cm thick interval of yellowish red calcareous

chalk (199.55–199.59 mbsf). Within this interval isa 1 cm-thick brownish yellow layer of calcareouschalk. Sediments in the yellowish red layers areweakly laminated and contain clay, dolomite,calcite and a minor percentage of nanofossils.Iron manganese oxides and opaque minerals areminor components. The brownish yellow chalklayer (199.59–199.68 mbsf) contains clay, dolomite,iron manganese oxide, volcanic glass and abun-dant small mottles. Nanofossils are rare in theselayers (Shipboard Scientific Party of Leg 199,2002).

Across the P/E transition, a total of 116 sedi-ment samples were analyzed for inorganic chemis-try and foraminiferal assemblage. Resolution isgenerally every 2 cm (∼1 ky) but only every 1 cmclose to the P/E transition.

ANALYTICAL PROCEDURE

For the foraminiferal analysis, sediment sampleswere washed in a 63 μm sieve and then driedat < 80°C. Samples were sieved further through a150 μm sieve, and divided into aliquots for pickingwith a microsplitter. For assemblage analysis,more than 200 specimens (>150 μm) were picked,identified, and counted. We follow the taxonomy ofCushman (1935, 1946, 1951), Ellis and Messina(1940), Tjalsma and Lohmann (1983), Loeblich andTappan (1987), Nomura (1991), Speijer (1994) andWidmark (1997).

Sediment aliquots for inorganic chemistry werecrushed to fine powder after drying at 40–50°C.Samples were decomposed according to followingprocedure: 100 mg of grained powder aliquotsample was weighed into a 50 mL Teflon beaker,which was placed on a hot plate. Then, 3 mL ofHClO4, 2.5 mL of HCl and HNO3 and 5 mL of HFwere added. After the sample was decomposedonce at 95°C for 10 h and twice at 140°C for 12 h,the solutions were diluted with 5 mL of HNO3 anddistilled MilliQ-water to adjust the total volume to100 mL. Then the prepared sample solution wasanalyzed by a SPS7800 inductively coupled plasmaatomic emission spectrometer (ICP-AES) andHP-4500 inductively coupled plasma mass spec-trometer (ICP-MS) analyzers at the GeologicalSurvey of Japan (GSJ) according to Maeda et al.(2002). Reference rock standards provided by theGSJ (JB-2, JA-2, JG-1A, JG-2, JR-1) as well asstandard solutions prepared from pure elementalstandard solutions were used for calibration. Rela-tive standard deviation of >3 ppm for major and

ODP Leg 199 Site Map

160°W 150° 140° 130°

10°

20°

30°N

0 500 1000

Site 1215

Site 1216

Site 1217

Site 1218Site 1219

Site 1220Site 1221

Site 1222

Site 40

Site 42

Site 43

Site 69 Site 70

Site 71

Site 72

Site 160Site 161

Site 162

Site 163

Site 573

Site 574

Site 575

Honolulu

Leg 199 sitesDSDP sites

Molokai F.Z.

Clipperton F.Z.

Clarion F.Z.

km

Fig. 1 Location map for Site 1220 during Ocean Drilling Program ofBathymetric map of Leg 199 (Shipboard Scientific Party, 2002).

Ocean acidification and foraminifera 3

© 2015 Wiley Publishing Asia Pty Ltd

Page 4: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

minor elements was less than 1% for contents inthe sediments while for trace elements it was lessthan 5%.

RESULTS

INORGANIC ELEMENTS IN THE SEDIMENTS

Contents of inorganic elements (Ca, Al, Mn, Fe andBa) are plotted as a function of stratigraphicdistance below seafloor (Table 1) in Figure 2. Con-tents of major and minor constituents are: 1.25–38.8 wt% (av. 27.6 wt%) for Ca; 0.22–5.49 wt% (av.1.43 wt%) for Al; 0.016–7.73 wt% (av. 0.38 wt%) forMn; 0.30–6.84 wt% (av. 1.53 wt%) for Fe; and21–26 000 ppm (av. 5900 ppm) for Ba.

Their abundance in the sediments changeddramatically across the P/E transition, where bio-genic carbonate is depleted while clay mineralswith high abundance of heavy metals areenriched. Based upon color of sediments and rela-tive abundance of each elements and correlationamong various elements, this section is dividedinto six parts: (i) light brown sediments (above199.38 mbsf); (ii) dark brown sediments (199.38–199.50 mbsf); (iii) black sediments (199.50–199.55mbsf); (iv) Red sediments (199.55–199.59 mbsf);(v) brown sediments (199.59–199.68 mbsf);and (vi) Pale brown sediments (below 199.68mbsf). The P/E boundary was identified at199.68 mbsf.

Of these, light brown and pale brown sedi-ments represent fairly normal sediments depos-ited during the late Paleocene and the earlyEocene in the Pacific, respectively. Generallyspeaking, light brown color sediments can beattributed to higher contents of Ca in carbonatewhile pale brown color sediments show less car-bonate content. Ca was largely absent in theother sediments (dark brown, black, red andbrown sediments between 199.38 and 199.68mbsf) (Periods 2, 3, 4, and 5). Following previousstudies of the P/E transition (Zachos et al. 2005),we attribute the absence to carbonate dissolution.Sedimentary Al content shows the oppositetrend: low abundance in light brown and palebrown sediments (above 199.38 mbsf and below199.68 mbsf) (Periods 1 and 6) and high abun-dance in the other sediments. Higher content ofMn, which occurs in the form of Mn oxide, corre-sponds to black and dark brown sediments. Feand Ba were also enriched in dark brown, black,red and brown sediments.

FAUNAL ANALYSIS OF BENTHIC FORAMINIFERA

We conducted a high resolution analysis of benthicforaminiferal abundance from 198.90–199.79 mbsfin order to constrain the P/E transition (Fig. 3,Table S1). The analysis showed dramatic changesin abundance and species assemblage throughoutthese samples (Fig. 3).

Although carbonate content was sufficientlyhigh and preserved planktonic and benthic fora-minifera in the light brown calcareous chalk of thebottom of the Eocene, the number of foraminiferaper gram chronologically increased from lessthan 100 to about 300 at around 199.18 mbsf. Thetaxa such as Nuttallides truempyi, Oridorsaliaumbonatus and Pleurostomella paleocenica werecharacterized by their small size. In this zone,Coryphostoma crenulata and Ellipsoglandulinavelascoensis occurred only in the Eocene section.These benthic foraminifers were very thin walled.In contrast, planktonic foraminifers were veryabundant, which commonly reached as high as90%.

Both benthic and planktonic foraminifersdecreased in abundance downcore. Planktonicforaminiferal tests became increasingly frag-mented in multicolored beds near the P/E transi-tion. Benthic foraminifers were extremely rare orcompletely absent in the dark brown, black,red and brown sediments (199.38–199.68 mbsf)(Periods 2, 3, 4, and 5). Nuttallides truempyi andO. umbonatus were very poorly preserved andshowed secondary calcite overgrowths on test sur-faces at 199.32–199.39 mbsf (light brown/darkbrown sediment transition). Both benthic andplanktonic foraminifers with normal calcareoustests were absent at 199.48–199.49 (dark brownsediment) in Period 2, 199.51–52 (dark brown sedi-ment) in Period 3, 199.54–59 (dark brown to redsediment) in Periods 3 and 4 and 199.62–199.63mbsf (brown sediment) in Period 5. These intervalsshowed the most marked color changes in this coreand were associated with well-developed dolomitecrystals.

Most interestingly, benthic foraminiferawith agglutinated walls (Saccorhiza ramosa,Rhabdammina sp., Ammoglobigerina sp. andHaplophragmoides sp.) became dominant inthese intervals in spite of the low overall numberof foraminifera per gram. A sample at 199.64–68mbsf (brown sediment) included poorly pre-served N. truempyi, O. umbonatus, Buliminatuxpamensis, Tappanina selmensis and Valva-labamina sp., all of which were heavily

4 H. Kawahata et al.

© 2015 Wiley Publishing Asia Pty Ltd

Page 5: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

Table 1 The contents of major and minor elements (Ca, Al, Mn, Fe and Ba) of sediments across the P/E boundary at Site1220. (data will be stored in a paleo-data bank, http://www.pangaea.de)

Cor Section Topcm

Bottomcm

Depth (mbsf)mbsf

Zone Cawt%

Alwt%

Mnwt%

Fewt%

Bappm

N1 20 X 1 15 17 197.55 1 32.85 0.47 0.103 0.83 5 759N2 20 X 1 18 20 197.58 1 35.62 0.34 0.061 0.51 4 926N3 20 X 1 21 23 197.61 1 35.43 0.33 0.030 0.47 4 514N4 20 X 1 24 26 197.64 1 36.33 0.35 0.078 0.61 5 163N5 20 X 1 27 29 197.67 1 35.98 0.33 0.069 0.57 4 444N6 20 X 1 30 32 197.70 1 35.64 0.43 0.073 0.57 4 305N7 20 X 1 33 35 197.73 1 31.60 0.39 0.101 0.81 4 362N8 20 X 1 36 38 197.76 1 34.77 0.36 0.064 0.53 4 793N9 20 X 1 39 41 197.79 1 36.95 0.31 0.063 0.42 4 112N10 20 X 1 42 44 197.82 1 35.41 0.40 0.071 0.54 5 365N11 20 X 1 45 47 197.85 1 32.71 0.36 0.066 0.52 4 823N12 20 X 1 48 50 197.88 1 35.69 0.40 0.064 0.51 4 822N13 20 X 1 51 53 197.91 1 35.77 0.37 0.055 0.46 4 753N14 20 X 1 54 56 197.94 1 37.46 0.37 0.055 0.44 4 444N15 20 X 1 57 59 197.97 1 36.38 0.34 0.055 0.48 3 940N16 20 X 1 60 62 198.00 1 36.77 0.27 0.043 0.39 3 657N17 20 X 1 63 65 198.03 1 37.33 0.27 0.047 0.38 3 572N18 20 X 1 66 68 198.06 1 38.16 0.31 0.058 0.42 3 846N19 20 X 1 69 71 198.09 1 38.33 0.34 0.060 0.49 4 460N20 20 X 1 72 74 198.12 1 38.08 0.31 0.042 0.45 3 995N21 20 X 1 75 77 198.15 1 37.99 0.28 0.047 0.35 3 750N22 20 X 1 78 80 198.18 1 37.60 0.29 0.046 0.38 3 712N23 20 X 1 81 83 198.21 1 36.85 0.31 0.055 0.40 4 282N24 20 X 1 84 86 198.24 1 37.92 0.29 0.049 0.39 3 785N25 20 X 1 87 89 198.27 1 37.58 0.22 0.037 0.30 3 218N26 20 X 1 90 92 198.30 1 38.00 0.24 0.042 0.40 3 227N27 20 X 1 93 95 198.33 1 33.65 0.38 0.083 0.72 3 885N28 20 X 1 96 98 198.36 1 34.34 0.45 0.093 0.76 5 082N29 20 X 1 99 101 198.39 1 36.55 0.28 0.056 0.41 4 645N30 20 X 1 102 104 198.42 1 36.46 0.31 0.047 0.41 4 129N31 20 X 1 105 107 198.45 1 33.29 0.44 0.088 0.65 5 606N32 20 X 1 108 110 198.48 1 34.28 0.35 0.060 0.50 5 138N33 20 X 1 111 113 198.51 1 34.26 0.45 0.063 0.53 5 505N34 20 X 1 114 116 198.54 1 35.54 0.39 0.082 0.61 5 527N35 20 X 1 117 118 198.57 1 34.15 0.43 0.073 0.59 5 141N36 20 X 1 118 120 198.58 1 36.26 0.42 0.099 0.58 6 149N37 20 X 1 121 123 198.61 1 37.18 0.26 0.050 0.34 4 464N38 20 X 1 124 126 198.64 1 35.53 0.34 0.065 0.45 5 470N39 20 X 1 127 129 198.67 1 36.01 0.37 0.072 0.55 4 780N40 20 X 1 130 132 198.70 1 36.08 0.36 0.076 0.51 5 141N41 20 X 1 133 135 198.73 1 38.07 0.30 0.065 0.45 4 608N42 20 X 1 136 138 198.76 1 35.53 0.51 0.119 0.76 6 153N43 20 X 1 139 141 198.79 1 38.55 0.39 0.075 0.52 5 073N44 20 X 1 142 144 198.82 1 38.78 0.39 0.078 0.53 4 623N45 20 X 1 145 147 198.85 1 36.76 0.38 0.091 0.60 5 517N46 20 X 2 0 2 198.90 1 37.92 0.35 0.078 0.60 4 732N47 20 X 2 2 4 198.92 1 37.24 0.34 0.081 0.56 4 748N48 20 X 2 4 6 198.94 1 37.42 0.35 0.070 0.52 4 856N49 20 X 2 6 8 198.96 1 37.82 0.36 0.073 0.54 4 664N50 20 X 2 8 10 198.98 1 37.19 0.40 0.078 0.57 4 724N51 20 X 2 10 12 199.00 1 36.99 0.42 0.087 0.66 5 397N52 20 X 2 12 14 199.02 1 36.66 0.44 0.087 0.65 5 368N53 20 X 2 14 16 199.04 1 34.95 0.51 0.091 0.67 6 148N54 20 X 2 16 18 199.06 1 36.63 0.48 0.078 0.63 4 663N55 20 X 2 18 20 199.08 1 35.78 0.54 0.079 0.70 3 126N56 20 X 2 20 22 199.10 1 36.36 0.48 0.059 0.66 919N57 20 X 2 22 24 199.12 1 37.90 0.45 0.045 0.48 374N58 20 X 2 24 26 199.14 1 35.65 0.55 0.064 0.59 667N59 20 X 2 26 28 199.16 1 35.22 0.56 0.072 0.66 1 591N60 20 X 2 28 30 199.18 1 35.78 0.54 0.053 0.58 426

Ocean acidification and foraminifera 5

© 2015 Wiley Publishing Asia Pty Ltd

Page 6: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

Table 1 Continued

Cor Section Topcm

Bottomcm

Depth (mbsf)mbsf

Zone Cawt%

Alwt%

Mnwt%

Fewt%

Bappm

N61 20 X 2 30 32 199.20 1 36.76 0.34 0.028 0.31 423N62 20 X 2 32 34 199.22 1 37.59 0.38 0.035 0.31 297N63 20 X 2 34 36 199.24 1 38.83 0.33 0.033 0.31 273N64 20 X 2 36 38 199.26 1 35.76 0.60 0.062 0.57 1 729N65 20 X 2 38 40 199.28 1 36.13 0.67 0.068 0.65 1 033N66 20 X 2 40 42 199.30 1 35.75 0.63 0.056 0.58 1 762N67 20 X 2 42 44 199.32 1 35.18 0.61 0.074 0.66 1 990N68 20 X 2 44 46 199.34 1 35.83 0.66 0.077 0.58 1 146N69 20 X 2 46 48 199.36 2 34.28 0.86 0.090 0.69 3 406N70 20 X 2 48 49 199.38 2 11.76 3.80 0.381 3.40 9 115N71 20 X 2 49 50 199.39 2 5.86 4.45 0.547 3.64 1 1 293N72 20 X 2 50 51 199.40 2 7.35 4.20 0.403 3.17 1 1 772N73 20 X 2 51 52 199.41 2 7.55 4.07 0.351 3.24 9 835N74 20 X 2 52 53 199.42 2 5.09 4.46 0.403 3.64 7 836N75 20 X 2 53 54 199.43 2 4.54 3.70 0.264 2.56 9 702N76 20 X 2 54 55 199.44 2 6.16 4.41 0.340 3.12 1 7 317N77 20 X 2 55 56 199.45 2 5.80 4.39 0.308 3.18 1 6 370N78 20 X 2 56 57 199.46 2 5.98 4.60 0.399 3.27 1 3 321N79 20 X 2 57 58 199.47 2 5.38 4.34 0.393 3.51 1 6 009N80 20 X 2 58 59 199.48 2 5.88 4.08 0.547 3.43 2 1 410N81 20 X 2 59 60 199.49 2 4.31 4.23 0.663 4.13 1 4 234N82 20 X 2 60 61 199.50 2 4.99 4.19 0.925 3.85 1 6 756N83 20 X 2 61 62 199.51 2 7.04 4.12 1.726 3.83 1 3 019N84 20 X 2 62 63 199.52 3 4.82 3.55 7.612 2.91 7 858N85 20 X 2 63 64 199.53 3 4.46 3.14 7.733 2.85 7 439N86 20 X 2 64 65 199.54 3 1.25 2.57 7.702 3.87 5 401N87 20 X 2 65 66 199.55 4 2.03 4.10 4.208 4.92 2 875N88 20 X 2 66 67 199.56 4 4.04 5.10 0.099 3.08 619N89 20 X 2 67 68 199.57 4 6.36 4.81 0.091 2.16 661N90 20 X 2 68 69 199.58 4 2.03 5.49 0.074 2.59 1 279N91 20 X 2 69 70 199.59 4 4.71 4.70 0.719 3.30 1 1 430N92 20 X 2 70 71 199.60 4 6.02 4.94 0.249 4.56 2 6 092N93 20 X 2 71 72 199.61 4 5.81 4.09 0.426 4.84 2 4 079N94 20 X 2 72 73 199.62 4 8.18 3.41 0.295 4.69 1 9 631N95 20 X 2 73 74 199.63 5 4.03 3.62 0.211 6.73 1 5 819N96 20 X 2 74 75 199.64 5 3.73 3.64 0.127 6.84 1 5 147N97 20 X 2 75 76 199.65 5 6.59 3.49 0.299 6.31 1 4 503N98 20 X 2 76 77 199.66 5 3.51 3.67 0.153 6.83 1 8 191N99 20 X 2 77 78 199.67 5 5.12 3.65 0.605 5.43 1 8 420N100 20 X 2 78 79 199.68 5 9.89 3.19 0.174 5.09 1 6 274N101 20 X 2 79 80 199.69 5 25.38 1.67 0.085 2.46 6 754N102 20 X 2 80 81 199.70 6 33.24 0.86 0.025 1.11 194N103 20 X 2 81 82 199.71 6 29.34 0.95 0.028 1.28 200N104 20 X 2 82 83 199.72 6 33.49 0.80 0.056 1.01 126N105 20 X 2 83 84 199.73 6 33.18 0.73 0.035 0.98 32N106 20 X 2 84 85 199.74 6 37.28 0.54 0.031 0.66 21N107 20 X 2 85 86 199.75 6 34.51 0.68 0.018 0.88 47N108 20 X 2 86 87 199.76 6 33.20 0.63 0.039 0.85 1 518N109 20 X 2 87 88 199.77 6 35.54 0.57 0.030 0.77 388N110 20 X 2 88 89 199.78 6 35.86 0.56 0.019 0.76 436N111 20 X 2 89 90 199.79 6 34.22 0.63 0.073 0.89 955N112 20 X 2 90 91 199.80 6 35.09 0.57 0.035 0.78 2 375N113 20 X 2 91 92 199.81 6 34.23 0.52 0.016 0.68 1 580N114 20 X 2 92 93 199.82 6 31.95 0.56 0.043 0.71 400N115 20 X 2 93 94 199.83 6 31.90 0.58 0.017 0.61 74N116 20 X 2 94 95 199.84 6 28.33 0.77 0.036 0.90 267

6 H. Kawahata et al.

© 2015 Wiley Publishing Asia Pty Ltd

Page 7: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

Zeo

lites

Cal

cite

Cal

cisp

here

s

Cla

y m

iner

als

For

amin

ifers

Dol

omite

Fe

and

Mn

oxid

es, o

paqu

e m

iner

als

Vol

cani

c gl

ass

Nan

nofo

ssils

Rar

e (<

5%)

Com

mon

(<

10%

)A

bund

ant (

<20

%)

setil oeZ

sl ar eni m euqapo dna sedi xo n M- eFssal g ci nacl oV

de R n wor B n wor b el ap yr eVkcal Bn wor b t hgi L n wor b kr a D

kl ahc li ssof onna N kl ahc suoer acl a C yal C kl ahc li ssof onna N

050

sli ssof onna Nci not knal Psr efi ni mar of

ci ht neBsr efi ni mar of

eti mol o D

deni ar g- eni Feti cl ac

100

Cal

c.ch

alk

65 c

m (

199.

55)

25 c

m (

199.

15)

33 c

m (

199.

23)

8 cm

(19

8.98

)

32 c

m (

199.

23)

44 c

m (

199.

34)

47.3

cm

(19

9.37

)48

cm

(19

9.38

)

54 c

m (

199.

44 )

59 c

m (

199.

49)

63 c

m (

199.

53)

68 c

m (

199.

58)

70 c

m (

199.

6)

85 c

m (

199.

75)

76 c

m (

199.

66)

66 c

m (

199.

56)

4 cm

(19

8.94

mbs

f)

0 10 20 30 40 50 60 70 80 90cm(%

)0

714

2128

3542

01

23

45

6

02

46

8

02

48

6

00.

51.

52

31

2.5

Ca

(wt%

)

Al (

wt%

)

Mn

(wt%

)

Fe

(wt%

)

Ba

(104

ppm

)

Ca

Al

Mn

Fe

Ba

199.

80

199.

60

199.

50

199.

40

199.

30

199.

20

199.

10

199.

00

198.

90

199.

70

mbs

fP/E

bo

unda

ry

Fig.

2Th

eco

nten

tsof

majo

rand

min

orele

men

ts(C

a,Al

,Mn,

Fean

dBa

)ofs

edim

ents

acro

ssth

eP/

Ebo

unda

ryat

Site

1220

.Pho

togr

aph

ofse

dim

ents

with

the

desc

riptio

nof

min

eralo

gyca

me

from

Ship

boar

dSc

ientifi

cPa

rty(2

002)

.The

P/E

boun

dary

was

iden

tified

at19

9.68

mbs

f.

Ocean acidification and foraminifera 7

© 2015 Wiley Publishing Asia Pty Ltd

Page 8: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

s i m r o f i i r a c c e b a l l e n i l e v a G

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

i e t i h w a n i l u c i t n e L

i t d n a r b e l l i h s e d i n o p e o e N

a l l e m o t s o l i t S . p s

s e d i o n i d i o r y G . p s

a i r a l u t x e T . p s

a n i l u d i s s a c o b o l G . p s

i y p m e u r t s e d i l l a t t u N

a t u c a e a r p s e d i o n i l a m o n A

s u t a n o b m u s i l a s r o d i r O

i g a o p . A / a t a r d a u q a n i m a s s y b A

s i s n e m a p x u t a n i m i l u B

a t a d i m a r y p a n i y r d u a G

a n i m i l u b e a r P . p s

s i s n e m l e s a n i n a p p a T

199.

68 m

bsf

6-13

13-2

0

21-2

8

28-3

5

35-4

3

42-5

0

50-5

7

57-6

4.5

65-7

2

74-7

8

77.5

-85

88-9

5 ) m c ( l a v r e t n I

199-

1220

B-2

0X-2

P/E

bo

unda

ry

For

amin

ifera

l tot

al n

umbe

r P

(w

t%)

Per

cent

ages

of

Agg

lutin

ated

ass

embl

age

(%)

199.

80

199.

60

199.

50

199.

40

199.

30

199.

20

199.

10

199.

00

198.

90

199.

70

mbs

f

0 10

0 20

0 30

0 40

0 50

0 0

20

40

80

100

60

0 20

40

80

10

0 60

0 10

0 20

0 30

0 40

0 50

0 0

20

40

80

100

60

0 20

40

80

10

0 60

(c

m)

90

70

60

50

40

30

20

10

0 80

Fig.

3Th

edi

strib

utio

nof

fora

min

ifera

lnum

berp

erse

dim

entg

ram

,per

cent

ages

ofag

glut

inate

das

sem

blag

esin

the

total

num

bero

fben

thic

fora

min

ifera

ndpl

ankto

nic

fora

min

ifera

perc

entag

ein

total

occu

rrenc

eof

fora

min

ifera

acro

ssth

eP/E

boun

dary

.Bas

edon

thel

asto

ccur

renc

eoft

hetax

aoft

heVe

lasco

-type

strati

grap

hicd

istrib

utio

nof

bent

hicf

oram

inife

rs,t

heP/

Ebo

unda

rywa

side

ntifi

ed(S

hipb

oard

Scien

tificP

arty,

2002

).

8 H. Kawahata et al.

© 2015 Wiley Publishing Asia Pty Ltd

Page 9: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

recrystallized. These samples contain abundantsmall (∼160–190 μm) crystals of dolomite, whosegrowth might account for the poor preservation ofthese species. Planktonic foraminifers were veryrare in these samples.

Members of the Velasco-type assemblage(i.e., deep-water benthic faunas) such as Gaveli-nella beccariiformis,Globorotalites micheliniana,Lenticulina whitei, Neoepnides hillebrandtiand Stilostomella plummerae were found in allanalyzed samples below 199.68 mbsf (Period 6).Based on last occurrence of the taxa, the P/E tran-sition is identified at a depth of 199.68 mbsf in thishole.

In the late Paleocene pale brown sediment(199.68–199.79 mbsf) in Period 6, carbonatecontent increased to high values for carbonaceoussediments. The number of foraminifera per gramranged from 133 to 220.

DISCUSSION

OXIC DEEP WATER ENVIRONMENTS AT THE P/ETRANSITION

Bralower et al. (1997) suggested that deep-wateroxygen deficiency is the main cause for a massextinction of deep-sea fauna during the P/E tran-sition, citing evidence for a major episode of explo-sive volcanism. Oxygen deficiency can be causedby the oxidation of substantial amounts ofmethane. However, in this study, we find no evi-dence of anoxia in the sediments around the P/Etransition at Site 1220 (Fig. 2). Black color sedi-ments are often attributed to ocean anoxic envi-ronments as is observed in black shales observedduring ocean anoxic events (OAE) I and II duringthe Cretaceous. These sediments often contain asmuch as 5 wt% of organic carbon (Takashima et al.2009). However, in core 1220, abundant Mn oxidesare responsible for the dark color in the black andred sediments (199.50–199.59 mbsf), where ferric(F3+) oxides were precipitated (Fig. 2). It indicatesa well-oxygenated deep-sea water across the P/Etransition.

We also find elevated content of Ba in the black,red and brown sediments (199.50–199.68 mbsf) atSite 1220 (Fig. 2). We take this as indication ofoxygenated waters as Bains et al. (2000) reportedthat barite is well preserved in oxic sedimentsfrom the equatorial to high latitude in the Atlanticat the P/E transition, which indicates a lack of Baremobilization due likely to sulfate reduction.Kennett and Stott (1991) also pointed out that deep

waters are unlikely to have become anoxic duringthe excursion because there is no increase inorganic carbon accumulation in the AntarcticOcean and possibly elsewhere.

Warm saline bottom water might be one offactors that lead to the mass extinction at the P/Etransition. Bice and Marotzke (2001) examinedwhether deep-water formation could haveoccurred in subtropical latitudes in the early Ceno-zoic using a global ocean general circulation modelforced by zonal mean SST and wind stressesderived from an atmospheric general circulationmodel (AGCM) simulation of the warm P/E tran-sition interval (∼55 Ma) and are held constant for aseries of sensitivity tests. In their simulationwhere subtropical evaporation increases and high-latitude precipitation is enhanced, a dramaticresponse is seen in the temperature and salinitystructure of the model ocean. However, the simu-lation does not result in deep convection in sub-tropical latitudes. In all cases, bottom water isformed in the southern high latitude. The Biceand Marotzke (2001) study therefore does notsupport the notion that warm saline bottom watercontributed to a mass extinction of deep-sea fauna.

Other lines of observation support the notionthat the deep ocean during the P/E transitionexperienced acidification, which may have con-tributed to the mass extinction at that time.Lower pH would be consistent with the wellknown P/E transition sediments marked by finelylayered deposits devoid of carbonate in severalsediment sections from the north and centralAtlantic Ocean (Dickens 2004). A leading expla-nation for the large negative δ13C shift in seawa-ter, as indicated by benthic foraminifera, duringthe P/E transition is the disintegration ofmethane hydrate. With a mean δ13C of <−60‰,methane can explain the negative shift. Uponrelease to the water column, methane wasexpected to be oxidized to carbon dioxide in theocean and atmosphere. CO2 is an acidic andtherefore it would lead to a reduction in deep-seapH, possibly triggering a rapid (<10 ky) shoalingof the CCD by >2 km, followed by gradualrecovery (>100 ky) (Zachos et al. 2005). A recentmodel-data comparison suggests further thatmethane by itself may have been insufficient tocause the expected deep sea acidification and thatan additional source of CO2 from volcanoes and/orthe oxidation of sedimentary organic carbon isrequired to account for the intensity of bottomcarbonate dissolution in the Atlantic and Pacific(Panchuk et al. 2008).

Ocean acidification and foraminifera 9

© 2015 Wiley Publishing Asia Pty Ltd

Page 10: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

OCCURRENCE OF AGGLUTINATED FAUNA IN THE DEEPSEA ACROSS THE P/E TRANSITION

Three important characteristics about the occur-rence of planktonic and benthic foraminifera aredemonstrated (Fig. 3): (i) Calcareous foraminiferaltests are quite poorly preserved at the P/E transi-tion. (ii) In spite of no calcareous benthic foramin-ifera, a few agglutinated benthic foraminiferasuch as S. ramosa, Rhabdammina sp. andAmmoglobigerina sp. occurred in the sedimentsin quite low numbers. (iii) Planktonic foraminiferatests were largely dissolved but some were persis-tent against dissolution on the seafloor.

The apparent change in the dominance of calcar-eous to agglutinated benthic fauna at the P/E tran-sition at Site 1220 is particularly noteworthy andindicates the effect of increased acidic condition inthe deep sea. The same feature of non-calcareousagglutinated foraminifera increase dramaticallywas observed, probably related to intense but nocomplete dissolution of carbonate across the P/Etransition in Egypt, Italy and Spain (Alegret &Ortiz 2006; Zili et al. 2009; Giusberti et al. 2009).The agglutinated species e.g., Glomospiracharoides, Karrerulina horrida, Rzehakinaepigone, Ammodiscus spp. and Gaudryinapyramidata, assumed to be restricted to deep-seapaleo-environments, constitute an important pro-portion of the benthic foraminiferal assemblages insouthern Tethys (Zili and Zaghbib–Turki 2010).These lines of evidence indicate that the aggluti-nated species contributed to benthic foraminiferamuch across the P/E transition.

In hydrothermal vents and mounds of theOkinawa Trough, bottom water temperature andsalinity anomalies are commonly small except inthe immediate vicinity of active hydrothermalchimneys. In the bottom and interstitial watersthat are more acidic than the surroundingambient bottom water, the proportion of theagglutinated foraminifera to total benthic fora-minifera (A/T) was generally high (>40%). Thisvalue decreases in a very predictable manner withincreasing distance from a hydrothermal vent.This observation indicates that the distinct eco-logical advantage that agglutinated foraminiferaenjoy in waters where acidity becomes stronger(Akimoto et al. 1990). Similar results of the domi-nance by agglutinated species due to acidificationwere reported in an active volcanic caldera(Finger & Lapps 1981). Organic-cemented agglu-tinated walls can be deposited in any waterregardless of corrosivity with respect to carbon-

ate. The majority of species live satisfactory onlyabove the CCD (Murray 2008). The sub-CCDfaunas of the modern Pacific are comprised ofdiverse agglutinated foraminifera (Saidova 1975).Also spatial community shift from hard to softcorals was reported in acidified water (Inoue et al.2013). These lines of evidence suggest that a pHdecrease was very likely responsible for the emer-gence of agglutinated foraminiferal fauna as cal-careous fauna was eliminated by acidification atthe P/E transition at Site 1220.

OCEAN ENVIRONMENTS AFFECTING PLANKTONIC ANDBENTHIC FORAMINIFERA AT THE P/E TRANSITION

The limited presence of tests of planktonic fora-minifera in the sediments of the P/E transition incore 1220 (Fig. 3) indicates that most of the testswere dissolved in the water column and in the sedi-ments. The absence of the more resistant calcare-ous benthic foraminifera and the presence insteadof the agglutinated foraminifera indicates that cal-careous benthic foraminifera suffered much andhardly lived on the seafloor at the P/E transition atSite 1220.

We do not necessarily imply that the carbonateion concentration of the deep waters at the timewas undersaturated with respect to minerals,calcite or aragonite. In fact the preservation ofplanktonic foraminiferal tests, even in smallamounts, would argue that the deep waters werenot undersaturated. We suggest, however, that thedegree of supersaturation was sufficiently low thatthe calcification of benthic foraminifera wasseverely retarded or limited and this is likely thecause of their extinction. The effect of the degreeof saturation on calcification rates has been dem-onstrated in recent years by a number of labora-tory and field studies. Published data on corals,coccolithophores and foraminifera all show a defi-nite reduction in calcification by 5–25% in responseto a doubling of pCO2 from pre-industrial values(from 280 to 560 ppm pCO2) (Feely et al. 2004).This reduced calcification rate also occurs evenwhen the carbonate saturation level was above one(at values below one, carbonate dissolved). A posi-tive correlation between foraminiferal shell weightand the carbonate saturation state is observed inthe palaeoceanographic record as a response toknown glacial-interglacial change in pCO2 of thepast 50 ky (Baker & Elderfield 2002). Plotting cal-cification rate versus the carbonate ion concentra-tion reveals a similar slope with one half to onethird decrease in calcification rate with [CO3

2−]

10 H. Kawahata et al.

© 2015 Wiley Publishing Asia Pty Ltd

Page 11: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

decreasing from 400 to 100 μM (Erez 2003). Inaddition, culture study of cloned specimens ofthe large benthic foraminifer, Marginoporakudakajimensis, indicate that their growth rate,measured by shell diameter, shell weight and thenumber of chambers added, decreased with lower-ing pH, lowering [CO3

2−], and shell weight, inparticular, was more closely dependent upon pHthan shell diameter or the number of chambers(Kuroyanagi et al. 2009; Hikami et al. 2011).Anomalous shift in foraminiferal test size acrossthe P/E transition could result from pH reductionalthough Kaiho et al. (2006) attributed it to defi-cient dissolved oxygen.

In the case of deep sea, pressure is also anotherimportant factor to control carbonate saturationlevel, which controls calcification rate and carbon-ate dissolution. Saturation level of [CO3

2−]decreases rapidly versus water depth if Ca concen-tration is constant. The values are 1.9 and 3.2 timesof that of surface water at a water depth of 3000 mand 6000 m, respectively, at temperature of 2°Cand salinity of 35 per (or permill). Therefore itwould be concluded that ocean acidification at deepsea with reduced carbonate saturation level wouldhave been the most plausible cause for drasticchange from calcareous to agglutinated benthicforaminiferal fauna at the P/E transition. It is sup-ported by the observation that benthic foramin-ifera from marginal and epicontinental basinsshow lesser extinctions and/or temporary assem-blage change (Alegret & Ortiz 2006).

DEEP AND SURFACE WATER CONDITION ACROSSTHE P/E TRANSITION

The atmospheric carbon dioxide concentrations inthe early Cenozoic era (about 45–60 Ma) are widelybelieved to have been much higher than at present.For example, boron-isotope ratios of planktonicforaminiferal shells may provide some constraintson estimation of the paleo-pH and indicate thatatmospheric pCO2 was perhaps >2000 ppm fromabout 60 to 52 Ma (Pearson & Palmer 2000). Thisvalue is, however, likely too high because thesevalues were comparable or higher than estimatedpCO2 even during the more typical ‘hot house’,Cretaceous times (Takashima et al. 2009). It hasfurther been pointed out that current boronisotope technique has considerable uncertaintybecause of unconstrained boron isotope budget inthe past ocean (Lemarchand et al. 2000; Paganiet al. 2005). In contrast, paleobotanical evidencesuggests that pCO2 remained between 300 and

450 ppm between 54 and 58 Ma with the exceptionof a single high estimate (∼800 ppm) near the P/Etransition (Royer et al. 2001). Assuming that theextinction of calcareous benthic foraminifera at theP/E transition is indeed due to deep ocean acidifi-cation as we suggested in this study, we canprovide some constraint on the atmospheric pCO2

during that time by using a simple three boxmodel, which is similar to that described byYamamura et al. (2007).

In Figure 4, we show how Δ[CO32−], the degree

of carbonate ion saturation, relative to calcitedepends on alkalinity and atmospheric pCO2 at thesurface (dotted lines) and 3000 m water depth(solid lines) for the modern (Fig. 4a) and P/Etransition (Fig. 4b). In making these figures, weassumed the same strength of the biological pump(see Fig. 4 caption for description of method). Forboth times, we can see that Δ[CO3

=] at the surfaceis about 200 μmol/kg higher than at 3000 m. This isthe dominant pressure effect on Δ[CO3

=]. Forexample, at the preindustrial atmospheric pCO2 of280 ppm and deep alkalinity of 2435 μmol-eq/kg,Δ[CO3

=] is 200 μmol/kg at the surface but almost0 μmol/kg at 3000 m for the modern condition (seeopen circle in Fig. 4a). This is a reasonable repre-sentation of the modern Pacific, where much of thedeep ocean is undersaturated. This thermody-namic pressure effect obviously exists for the P/Econditions, where the same preindustrial pCO2 andalkalinity value give Δ[CO3

=] of about 240 μmol/kgat the surface and 40 μmol/kg at the surface (seeopen circle in Fig. 4b). The surface-deep differenceis still about 200 μmol/kg. The difference of about40 μmol/kg at 3000 m and depth between themodern and P/E conditions is due to the highertemperature at the P/E time. This is a direct con-sequence of the fact that the higher SST reducesthe CO2 solubility and therefore there is less dis-solved inorganic carbon (DIC) in the warmer P/Eocean. For the same atmospheric pCO2 and meanocean alkalinity, the world ocean with less DIC ismore basic and Δ[CO3

=] is higher. This means thatin order to make the P/E deep ocean undersatu-rated or close to Δ[CO3

=] = 0, the atmosphericpCO2 must be significantly higher (or alkalinitymuch lower). Figure 4 shows that for the samemean alkalinity as today, pCO2 must increase from280 ppm to >410 ppm in order to make the P/Edeep ocean undersaturated (see the Δ[CO3

=]= 0contour line in Fig. 4b).

If the ocean mean alkalinity were much higher,then the required pCO2 to the P/E ocean under-saturated would be even higher. There is indeed

Ocean acidification and foraminifera 11

© 2015 Wiley Publishing Asia Pty Ltd

Page 12: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

expectation that the P/E ocean had a higher alka-linity, if we assume that the CH4 released was oxi-dized to CO2 and dissolved deep-sea carbonate. Astotal of 2000 peta gC of CH4 was converted to CO2

and reacted with calcium carbonate in theocean, alkalinity may increase by a maximum of200 μmol-eq/kg, which corresponds to approxi-mately 10% increase in alkalinity in the ocean. Ofcourse, duration time is estimated to be 10 ky.Alkalinity increase at an initial stage might bean order of 100 μmol-eq/kg. Higher alkalinityalso raises the oversaturation level. However,Figure 4b shows that the deep ocean becomesundersaturated at pCO2 of about 530 and 470 ppmat the maximum and a half of the alkalinity range,respectively. As mentioned before, at the P/E tran-

sition, the pCO2 suddenly increased by severalhundred ppm or more, which would be consistentwith the deep-sea acidification for the dissolutionof inorganic carbonate and much difficulty of theproduction of calcareous shells of benthic foramini-fer. In contrast, surface ocean is still oversatura-tion of calcium carbonate due to the pressure effect(Fig. 4b).

IMPLICATION FOR FUTURE ENVIRONMENTS

Based upon mass-balance calculations, previousstudies estimated that a total amount of 2000 petagC of CH4 in oceanic hydrates (with a δ 13C of−60‰) was rapidly released to the ocean andatmosphere at an average rate of ∼0.2 peta gC/y ofCO2 over 10 ky in order to explain the observed δ13C excursion at the P/E transition (Dickens et al.1995; Zachos et al. 2005). This rate is approxi-mately 1/30 of current global emission rate (∼7.3peta gC/y) of anthropogenic carbon input on timescales. This makes the P/E transition event is apossible analog of the modern human perturbationto the ocean-atmosphere carbon system (Dickenset al. 1995).

The rising level of pCO2 is making the world’soceans more acidic. The pH dropped from about

200 300 400 500 600 700 8002300

2350

2400

2450

2500

2550

2600

2650

2700

2750

2800

80

120

120

120

160

160

200

200

240

240

280

−40

−40

−40

0

0

40

4080

Atmospheric pCO2 (ppm)

Mea

n oc

ean

ALK

(um

ol-eq/

kg)

(a) Modern Δ[CO3=] at surface (z=0, T=22.1) and deep (z=3000, T=1.8)

200 300 400 500 600 700 8002300

2350

2400

2450

2500

2550

2600

2650

2700

2750

2800

120

120

160

160

160

200

200

24028

0

280

320

−40

0

0

0

40

40

80

8012

0160

Atmospheric pCO2 (ppm)

Mea

n oc

ean

ALK

(um

ol-e

q/kg

)

(b) P/E Δ[CO3=] at surface (z=0, T=28) and deep (z=3000, T=18)

Fig. 4 The dependence of the degree of carbonate ion saturation withrespect to mineral calcite at the surface (dotted lines) and deep ocean(solid lines) as a function of atmospheric pCO2 and mean alkalinity for themodern (a) and P/E transition (b). For any pressure or depth, carbonateion concentration can be readily calculated given four parameters: tem-perature, salinity, alkalinity and dissolved inorganic carbon (DIC). Salinityis assumed to be constant at 35 PSU for both times. For the moderntemperature, we used global mean annual temperature of 22.1°C at thesurface and 1.8°C at 3000 m for the modern time (Matsumoto et al.2002). For the P/E transition, we assumed 28°C for the surface and 18°Cat depth, respectively (Kennett & Stott 1991; Tripati & Elderfield 2004). Noassumption is made about the mean ocean alkalinity (vertical axis). Forboth times though, we assume the same carbonate pump strength astoday, so that the alkalinity concentration at the surface is set to be lowerby 113 μmol-eq/kg compared to the ocean mean (Matsumoto et al.2002). For the surface, we calculate the equilibrium concentrations of DICand carbonate ion using 35 PSU, assumed temperatures and for theranges of alkalinity and atmospheric pCO2 (horizontal axis). The deepocean DIC is obtained from surface ocean DIC assuming the moderncarbon pump strength as indicated by the deep-to-surface DIC gradient of375 μmol-eq/kg (Matsumoto et al. 2002). After subtracting the saturationcarbonate ion concentration from the estimated carbonate ion concentra-tion at the two pressure levels, the saturation carbonate ion concentrationor Δ[CO3

=] is obtained and plotted. The preindustrial values for oceanmean alkalinity of 2435 μmol-eq/kg and pCO2 of 280 ppm are indicatedby open circles. The indirect method of obtaining the deep DIC introducesan error of about 10%. This was determined from comparing Δ[CO3

=]calculated by using the indirectly obtained deep DIC and by using the insitu of deep DIC.◀

12 H. Kawahata et al.

© 2015 Wiley Publishing Asia Pty Ltd

Page 13: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

8.15 in the pre-industrial time to about 8.06 todayand is projected to further drop to 7.9 by the end ofthis century (Kleypas et al. 1999). This acidifica-tion will in turn reduced the degree of carbonateion saturation for the global ocean. Modern-daysurface [CO3

2−] varies meridionally by more than afactor of two, from average concentrations in theSouthern Ocean of 105 μmol/kg to average concen-trations in tropical waters of 240 μmol/kg. This low[CO3

2−] is due to (i) low SST and CO2-system ther-modynamics and (ii) large amounts of upwelleddeep water, which is more acidic from organicmatter remineralization. Orr et al. (2005) suggestthat the entire Southern Ocean and the subarcticPacific will be undersaturated with respect to ara-gonite by 2100. These severe saturation statesare expected to manifest first in winter, when thepCO2 in seawater is the highest at low temperatureand wind-driven mixing of subsurface water intothe mixed layer (Feely et al. 2004).

Some of these cold, low pH surface waters inSouthern Ocean will be exported northward as Ant-arctic Bottom Water into the Pacific, Atlantic andIndian Oceans. Large areas of deep seafloor(∼>4 km), especially in the North Pacific, willreceive more severe influence because of no carbon-ate sediments below CCD and/or little carbonatebetween lysocline and CCD. Of course, the atmo-spheric concentration of CO2 have various influenceto the ocean: acidification, SST increase, morestratified surface water, slowdown of ocean con-veyor belt, modification of dissolved oxygen indeep seawater, reduction of primary production,decrease in export production. As demonstrated inthe P/E transition, the co-variation will providemore severe environments to the calcareous benthicanimals (Thomas 2007). Kuroyanagi et al. (2013)cultured shallow-water benthic foraminifera livingin coral reefs. The high tolerance of the foraminiferato extremely low dissolved oxygen condition.

Of these, ocean acidification will probably be thefirst grade of factors to control benthic calcareousfauna (Dayton 1990; Freiwald et al. 2004; Greenet al. 2004). In order to assess future impact on theocean acidification, laboratory experiments andfield observations are required because of limiteddata at the moment.

SUMMARY AND CONCLUSIONS

ODP Leg 199 Site 1220 provides a continuous sedi-mentary section across the P/E transition in thecarbonate-bearing sediments on 56–57 Ma oceaniccrust.

1 The apparent change in the dominance of calcar-eous to agglutinated benthic fauna at the P/Etransition at Site 1220 is particularly notewor-thy, consistent with those obtained from TethysOcean and is very likely due to a pH decreaseunder well-oxygenated deep water.

2 The absence of the more resistant calcareousbenthic foraminifera and little or no preserva-tion of planktonic foraminifera indicates that nocalcareous benthic foraminifera lived on the sea-floor at the P/E transition at Site 1220. Highpressure influencing on the carbonate saturationis another important factor to control carbonatesaturation level. Saturation level of [CO3

2−]increases rapidly versus water depth.

3 Box model calculation demonstrates that, for thesame mean alkalinity as today, pCO2 mustincrease from 280 ppm to about 410 ppm inorder to make the P/E deep ocean undersatu-rated under oversaturated surface ocean. Anincrease in pCO2 would be consistent with thatestimated from paleobotanical evidence.

4 Current global emission rate (∼7.3 peta gC/y) ofanthropogenic carbon input is approximately 30times the estimate of that at the P/E transition.The benthic fauna will be threatened in combi-nation with ocean acidification, SST increase andmore stratified surface water in future.

ACKNOWLEDGEMENTS

We thank K. Minoshima and N. Hokanishi for theirtechnical support for inorganic chemical analysisat Geological Survey of Japan, National Instituteof Advanced Industrial Science and Technology.This research was partly supported by Grants-in-Aids from the Japan Society for the Promotion ofScience to H.K. (No. 22224009). We also acknowl-edge the constructive comments by Prof. MichaelA. Kaminski and one anonymous reviewer.

REFERENCES

AKIMOTO K., TANAKA T., HATTORI M. & HOTTA H. 1990.Recent benthic foraminiferal assemblages aroundhydrothermal vents in the Okinawa Trough, RyukyuIslands, Japan. In Takayanagi Y. and Saito T. (eds.)Studies in Benthic; Proceedings of the Fourth Inter-national Symposium on Benthic Foraminifera,Sendai, pp. 211–25, Tokai University Press.

ALEGRET L. & ORTIZ S. 2006. Global extinction event inbenthic foraminifers across the Paleocene/Eocene atthe Dababiya Stratotype section. Micropaleontology52, 433–47.

Ocean acidification and foraminifera 13

© 2015 Wiley Publishing Asia Pty Ltd

Page 14: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

BAINS S., NORRIS R. D., CORFIELD R. M. & FAUL K. L.2000. Termination of global warmth at thePalaeocene/Eocene boundary through productivityfeedback. Nature 407, 171–4.

BAKER S. & ELDERFIELD H. 2002. Foraminiferal calci-fication response to glacial-interglacial changes inatmospheric CO2. Science 297, 833–83.

BICE K. L. & MAROTZKE J. 2001. Numerical evidenceagainst reversed thermohaline circulation in thewarm Paleocene/Eocene ocean. Journal of Geo-physical Research 106, 11529–42.

BRALOWER T. J. 2002. Evidence of surface wateroligotrophy during the Paleocene-Eocene thermalmaximum: Nannofossil assemblage data from OceanDrilling Program Site 690, Maud Rise, Weddell Sea.Paleoceanography 17, 13-1-13-29.

BRALOWER T. J., THOMAS D. J., ZACHOS J. C. et al. 1997.High-resolution records of the late Paleocenethermal maximum and circum-Caribbean volcanism:Is there a causal link? Geology 25, 963–6.

BRALOWER T. J., ZACHOS J. C., THOMAS E. et al. 1995.Late Paleocene to Eocene paleoceanography of theequatorial Pacific Ocean: Stable isotopes recorded atODP Site 865, Allison Guyot. Paleoceanography 10,841–65.

CUSHMAN J. A. 1935. Upper Eocene foraminifera of thesoutheastern United States. Geological Survey Pro-fessional Paper United States 181, 1–88.

CUSHMAN J. A. 1946. Upper Cretaceous foraminifera ofthe Gulf Coastal region of the United States andadjacent areas. Geological Survey ProfessionalPaper United States 206, 1–241.

CUSHMAN J. A. 1951. Paleocene foraminifera of the GulfCoastal region of the United States and adjacentareas. Geological Survey Professional Paper UnitedStates 232, 1–75.

DAYTON P. K. 1990. Polar benthos. In Smith W. O. (ed.)Polar Oceanography, Part B: Chemistry, Biologyand Geology, pp. 631–85, Academic, San Diego.

DICKENS G. R. 2004. Hydrocarbon-driven warming.Nature 429, 513–5.

DICKENS G. R., CASTILLO M. M. & WALKER J. G. C.1997. A blast of gas in the latest Paleocene: Simulat-ing first-order effects of massive dissociation ofoceanic methane hydrate. Geology 25, 259–62.

DICKENS G. R., ONEIL J. R., REA D. K. & OWEN R. M.1995. Dissociation of oceanic methane hydrate as acause of the carbon-isotope excursion at the end ofPaleogene. Paleoceanography 10, 965–71.

ELLIS B. F. & MESSINA A. 1940. Catalogue of Foramin-ifera. Spec. Publ., Am. Mus. Nat. Hist. (Supplements,post-1940).

ENGEBRETSON D. C., COX A. & GORDON R. G. 1985.Relative motions between oceanic and continentalplates in the Pacific Basin. Special Paper GeologicalSociety of America 206, 1–60.

EREZ J. 2003. The sources of ions for biomineraliz-ation in foraminifera and their implications for

paleoceanographic proxies. In Dove P. M., De YoreoJ. J. and Weiner S. (eds.) Biomineralization Reviewsin Mineralogy and Geochemistry 54, pp. 115–49.Mineralogical Society of America, Blacksburg, Vir-ginia.

ERNST S. R., GUASTI E., DUPUIS C. & SPEIJER R. P.2006. Environmental perturbation in the southernTethys across the Paleocene/Eocene boundary(Dababiya, Egypt): Foraminiferal and clay mineralrecords. Marine Micropaleontology 60, 89–111.

FEELY R. A., SABINE C. L., LEE K. et al. 2004. Impact ofanthropogenic CO2 on the CaCO3 system in theoceans. Science 305, 362–6.

FINGER K. L. & LAPPS J. H. 1981. Foraminiferal deci-mation and repopulation in an active volcanic caldera,Deception Island, Antarctica. Micropaleontology 27,111–39.

FREIWALD A., FOSSA J. H., GREHAN A., KOSLOW T. &ROBERTS J. M. 2004. Cold-water Coral Reefs: Out ofSight – No Longer Out of Mind No. 22 in BiodiversitySeries, UNEP-WCMC, Cambridge, UK.

GIUSBERTI L., COCCIONI R., SPOVIERI M. & TATEO F.2009. Perturbation at the sea floor during themaximum: Evidence from benthic foraminifera atContessa Road, Italy. Marine Micropaleontology 70,102–19.

GREEN M. A., JONES M. E., BOUDREAU C. L., MOORE P.L. & WESTMAN B. A. 2004. Dissolution mortality ofjuvenile bivalves in coastal marine deposits. Limnol-ogy and Oceanography 49, 727–34.

GRIPP A. E. & GORDON R. G. 1990. Current plate veloci-ties relative to the hotspots incorporating theNUVEL-1 global plate motion model. GeophysicalResearch Letters 17, 1109–12.

HIKAMI M., FUJITA K., KUROYANAGI A. et al. 2011. Con-trasting ocean acidification responses of calcificationbetween two coral reef benthic foraminiferal speciesMarginopora kudakajimensis and Calcarinagaudichaudii. Geophysical Research Letters 38,doi:10.1029/2011GL048501.

INOUE S., KAYANNE H., YAMAMOTO S. & KURIHARA H.2013. Spatial community shift from hard to softcorals in acidified water. Nature Climate. Change 3,683–7.

JOHN C. M., BOHATY S. M., ZACHOS J. C. et al.2008. North American continental margin recordsof the Paleocene-Eocene thermal maximum:Implications for global carbon and hydrologicalcycling. Paleoceanography 23, doi: 10.1029/2007PA001465.

KAIHO K., TAKEDA K., PETRIZZO M. R. & ZACHOS J. C.2006. Anomalous shifts in tropical Pacific planktonicand benthic foraminiferal test size during thePaleocene-Eocene thermal maximum. Palaeogeo-graphy, Palaeoclimatology, Palaeoecology 237, 456–64.

KATZ M. E., PAK D. K., DICKENS G. R. & MILLER K. G.1999. The source and fate of massive carbon input

14 H. Kawahata et al.

© 2015 Wiley Publishing Asia Pty Ltd

Page 15: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

during the latest Paleocene thermal maximum.Science 286, 1531–3.

KELLY D. C., BRALOWER T. J., ZACHOS J. C., SILVA I. P.& THOMAS E. 1996. Rapid diversification of plank-tonic foraminifera in the tropical Pacific (ODP Site865) during the late Paleocene thermal maximum.Geology 24, 423–6.

KENNETT J. P. & STOTT L. D. 1991. Abrupt deep-seawarming, palaeoceanographic changes and benthicextinctions at the end of the Palaeocene. Nature 353,225 – 9.

KENT D. V., CRAMER B. S., LANCI L., WANG D., WRIGHT

J. D. & VAN DER VOO R. 2003. A case for a cometimpact trigger for the Paleocene/Eocene thermalmaximum and carbon isotope excursion. Earth Plan-etary Science Letters 211, 13–26.

KLEYPAS J. A., BUDDEMEIER R. W., ARCHER D.,GATTUSO J. P., LANGDON C. & OPDYKE B. N.1999. Geochemical consequences of increased atmo-spheric carbon dioxide on coral reefs. Science 284,118–20.

KOCH P. L., ZACHOS J. C. & GINGERICH P. D. 1992.Correlation between isotope records in marine andcontinental carbon reservoirs near the PaleoceneEocene boundary. Nature 358, 319–22.

KROON D., ZACHOS J. & BLUM P. 2004. Proceedings ofthe Ocean Drilling Program, Initial Report. 208,Ocean Drilling Program, College Station, TX.

KUROYANAGI A., DA ROCHA R. E., BIJMA J. et al. 2013.Effect of dissolved oxygen concentration on plank-tonic foraminifera through laboratory culture experi-ments and implications for oceanic anoxic events.Marine Micropaleontology 101, 28–32.

KUROYANAGI A., KAWAHATA H., SUZUKI A. & FUJITA K.2009. Impacts of ocean acidification on large benthicforaminifers: Results from laboratory experiments.Marine Micropaleontology 73, 190–5.

LEMARCHAND D., GAILLARDET J., LEWIN E. &ALLEGRE C. J. 2000. The influence of rivers onmarine boron isotopes and implications for recon-structing past ocean pH. Nature 408, 951–4.

LOEBLICH A. R. & TAPPAN H. 1987. ForaminiferalGenera and Their Classification. Van NostrandReinhold Co., New York.

MAEDA L., KAWAHATA H. & NOHARA M. 2002. Fluctua-tion of biogenic and abiogenic sedimentation on theShatsky Rise in the western North Pacific during thelate Quaternary. Marine Geology 189, 197–214.

MATSUMOTO K., SARMIENTO J. L. & BRZEZINSKI M. A.2002. Silicic acid leakage from the Southern Ocean asa possible mechanism for explaining glacial atmo-spheric pCO2. Global Biogeochemical Cycles, 16 (3).doi:10.1029/2001GB001442.

MURRAY J. W. 2008. Ecology and Applications ofBenthic Foraminifera. Cambridge University Press,Cambridge, U.K. p. 440.

NOMURA R. 1991. Paleoceanography of upperMaestrichtian to Eocene benthic foraminiferal

assemblages at Sites 752, 753, and 754, EasternIndian Ocean. In Weissel, J., Peirce, J., Taylor, E.,Alt, J., et al., Proceedings of the Ocean DrillingProgram, Scientific Results, 121, College Station, TX(Ocean Drilling Program), 3–30.

NORRIS R. D. & RÖHL U. 1999. Carbon cycling andchronology of climate warming during thePalaeocene/Eocene transition. Nature 401, 775–8.

NUNES F. & NORRIS R. D. 2005. Data report: High-resolution stable isotope records across thePaleocene/Eocene boundary, ODP sites 1220 and1221. In Wilson P. A., Lyle M. and Firth J. V. (eds.)Proceedings of the Ocean Drilling Program, Scien-tific Results, 199 MS number 199SR-206.

ORR J. C., FABRY V. J., AUMONT O. et al. 2005. Anthro-pogenic ocean acidification over the twenty-firstcentury and its impact on calcifying organisms.Nature 437, 681–6.

PAGANI M., LEMARCHAND D., SPIVACK A. &GAILLARDET J. 2005. A critical evaluation of theboron isotope-pH proxy: The accuracy of ancientocean pH estimates. Geochimica et CosmochimicaActa 69, 953–61.

PAGANI M., PEDENTCHOUK N., HUBER M. et al., THE

EXPEDITION SCIENTISTS 2006. Arctic hydrologyduring global warming at the Palaeocene/Eocenethermal maximum. Nature 442, 671–5.

PAK D., MILLER K. G. & BROWNING J. 1997. Globalsignificance of an isotopic record from the NewJersey Coastal Plain: Linkage between the shelf anddeep sea in the late Paleocene to early Eocene. InMiller, K. G. and Snyder, S. W. (eds.), Proc. ODP,Sci. Results, 150X: College Station, TX (OceanDrilling Program), 305–316. doi:10.2973/odp.proc.sr.150x.319.1997.

PANCHUK K., RIDGWELL A. & KUMP L. R. 2008. Sedi-mentary response to Paleocene-Eocene thermalMaximum carbon release: A model-data comparison.Geology 36, 315–8.

PEARSON P. N. & PALMER M. R. 2000. Middle Eoceneseawater pH and atmospheric carbon dioxide con-centrations. Science 284, 1824–6.

RAFFI I., BACKMAN J. & PÄLIKE H. 2005. Changes incalcareous nanofossil assemblages across thePaleocene/Eocene transition from the paleo-equatorial Pacific Ocean. Palaeogeography,Palaeoclimatology, Palaeoecology 226, 93–126.

RÖHL U., BRALOWER T. J., NORRIS R. D. & WEFER G.2000. New chronology for the late Paleocene thermalmaximum and its environmental implications.Geology 28, 927–30.

ROYER D. L., WING S. C., BEERLING D. J. et al. 2001.Palaeobotanical evidence for near present-day levelsof atmospheric CO2 during part of the Tertiary.Science 292, 2310–3.

SAIDOVA K. M. 1975. Benthonic Foraminifera of thePacific Ocean (in Russian). Akad. Nauk SSSR, Inst.Okeanol. Im. P. P. Shirshova, Moscow.

Ocean acidification and foraminifera 15

© 2015 Wiley Publishing Asia Pty Ltd

Page 16: Research Article Linkage of deep sea rapid acidification ......nanofossil genus Rhomboaster, the extinction of the nanofossil genus Fasciculithus, and the pres- ence of excursion

SCHEIBNER C. & SPEIJER R. P. 2008. Late Paleocene–early Eocene Tethyan carbonate platform evolution –A r. esponse to long- and short-term paleoclimaticchange. Earth and Science Reviews 90, 71–102.

SHIPBOARD SCIENTIFIC PARTY LEG 1999 2002. Leg 199summary. In Lyle M. W., Wilson P. A., Janecek T. R.et al. (eds.) Proceedings of the Ocean DrillingProgram, Initial. Reports., 199, pp. 1–87, U.S. Gov-ernment Printing Office, College Station TX (OceanDrilling Program)

SLUIJS A., SCHOUTEN S., PAGANI M. et al. 2006. TheExpedition 302 Scientists, Subtropical Arctic Oceantemperatures during the Palaeocene/Eocene thermalmaximum. Nature 441, 610–3.

SPEIJER R. P. 1994. Extinction and recovery patterns inbenthic foraminiferal paleo-communities across theCretaceous/Paleogene and Paleocene/Eocene bound-aries. Geologica Ultraiectina 124, 1–91.

STASSEN P., THOMAS E. & SPEIJER R. P. 2012. Inte-grated stratigraphy of the Paleocene-EoceneThermal Maximum in the New Jersey Coastal Plain:Towards understanding the effects of globalwarming in a shelf environment. Paleoceanography27, PA4210. 1-17.

STOREY M., DUNCAN R. A. & SWISHER III, C. C. 2007.Paleocene-Eocene Thermal Maximum and theopening of the Northeast Atlantic. Science 316,587–9.

SVENSEN H., PLANKE S., MALTHE-SORENSSEN A. et al.2004. Release of methane from a volcanic basin as amechanism for initial Eocene global warming.Nature 429, 542–5.

TAKASHIMA R., NISHI H., HAYASHI K. et al. 2009. Litho-,bio- and chemostratigraphy across the Cenomanian/Turonian boundary (OAE 2) in the VocontianBasin of southeastern France. PalaeogeographyPalaeoclimatology Palaeoecology 273, 61–74.

THOMAS E. 2007. Cenozoic mass extinctions in the deepsea; what disturbs the largest habitat on Earth? InMonechi S., Coccioni R. and Rampino M. (eds.) LargeEcosystem Perturbations: Causes and Conse-quences, pp. 1–24, Geological Society of America,Special Paper 424.

THOMAS E. & SHACKLETON N. J. 1996. The latest Paleo-cene benthic foraminiferal extinction and stableisotope anomalies. In Knox R. O. et al (ed.) Correla-tion of the Early Paleogene in Northwest Europe,pp. 401–41, Geological Society of London, SpecialPublication 101.

TJALSMA R. C. & LOHMANN G. P. 1983. Paleocene–Eocene bathyal and abyssal benthic foraminiferafrom the Atlantic Ocean. Micropaleontology Spec.Publ. 4, 1–90.

TRIPATI A. K. & ELDERFIELD H. 2004. Abrupt hydro-graphic changes in the equatorial Pacific and sub-tropical Atlantic from foraminiferal Mg/Ca indicategreenhouse origin for the thermal maximum at thePaleocene-Eocene Boundary. Geochemistry Geo-physics Geosystems 5, Q02006. doi:10.1029/2003GC000631.

WIDMARK J. G. V. 1997. Deep-sea benthic foraminiferafrom Cretaceous–Tertiary boundary strata in theSouth Atlantic Ocean: Taxonomy and paleoecology.Fossils Strata 43, 1–94.

XU W. Y., LOWELL R. P. & PELTZER E. 2001. Effect ofseafloor temperature and pressure variations onmethane flux from a gas hydrate layer: Comparisonbetween current and late Paleocene climate condi-tions. Journal Geophysical Research 106, 26413–23.

YAMAMURA M., KAWAHATA H., MATSUMOTO K.,TAKASHIMA R. & NISHI H. 2007. Paleoceanographyof the northwestern Pacific during the Albian.Palaeogeography, Palaeoclimatology, Palaeoecology254, 477–91.

ZACHOS J., PAGANI M., SLOAN L., THOMAS E. &BILLUPS K. 2001. Trends, rhythms, and aberrationsin global climate 65 Ma to present. Science 292, 686–93.

ZACHOS J. C., ROHL U., SCHELLENBERG S. A. et al. 2005.Rapid acidification of the ocean during the Paleocene-Eocene thermal maximum. Science 308, 1611–5.

ZILI L. & ZAGHBIB–TURKI D. 2010. Foraminiferalbiostratigraphy and palaeoenvironmental recon-struction of the Paleocene–Eocene transition at theKharrouba section, Tunisia (Southern TethysMargin). Turkish Journal Earth Science 19, 385–408.

ZILI L., ZAGHBIB–TURKI D., ALEGRET L., ARENILLAS

I. & MOLINA E. 2009. Foraminiferal turnover acrossthe Paleocene/Eocene boundary at the Zumayasection, Spain: Record of a bathyal gradual massextinction. Revista mexicana de ciencias geológicas26, 729–44.

SUPPORTING INFORMATION

Additional Supporting Information may be foundin the online version of this article at the publish-er’s web-site:

Table S1 The stratigraphic distribution of benthicforaminifer with sampled core depth and age atSite 1220. (data will be stored in a paleo-data bank,http://www.pangaea.de).

16 H. Kawahata et al.

© 2015 Wiley Publishing Asia Pty Ltd