21. Quantitative Distribution Patterns and Biomagnetostratigraphy of ...

Pisias, N.G., Mayer, L.A., Janecek, T.R., Palmer-Julson, A., and van Andel, T.H. (Eds.), 1995 Proceedings of the Ocean Drilling Program, Scientific Results, Vol. 138

21. QUANTITATIVE DISTRIBUTION PATTERNS AND BIOMAGNETOSTRATIGRAPHY OF MIDDLE AND LATE MIOCENE CALCAREOUS NANNOFOSSILS FROM EQUATORIAL INDIAN AND PACIFIC OCEANS (LEGS 115, 130, AND 138)1 Isabella Raffi,2 Domenico Rio,3 Anna d'Atri,4 Eliana Fornaciari,3 and Silvana Rocchetti3

ABSTRACT Selected calcareous nannofossils were investigated by means of quantitative methods in middle and upper Miocene sediments from the tropical Indian Ocean (ODP Leg 115) and equatorial Pacific Ocean (DSDPLeg 85, ODP Legs 130 and 138). Our goal was to test the reliability of the classic biohorizons used in the standard zonations of Martini (1971) and Bukry (1973) and, possibly, to improve biostratigraphic resolution in the Miocene. In a time interval of about 8 m.y., from the last occurrence (LO) of S. heteromorphus ( 13.6 Ma) to the LO of D. quinqueramus ( 5.5 Ma), a total 37 events were investigated, using both the conventional and some additional markers proposed in the literature. At least 17 of these events proved to be distinct biostratigraphic correlation lines between the two considered areas. This integrated biostratigraphic framework increases the biostratigraphic resolution in the middle-upper Miocene interval (of the order of about 0.5 m.y). All the investigated events were tied to the geomagnetic polarity time scale (GPTS) and compared to biomagnetostratigraphy from mid-latitude North Atlantic Site 94-608 (Olafsson, 1991; Gartner, 1992), thus obtaining further information about the biostratigraphic and biochronologic reliability of the investigated events and a significant improvement of the available nannofossil biomagnetostratigraphic model for the middle and late Miocene.

INTRODUCTION Recently, considerable efforts have been dedicated to improving, by means of calcareous nannofossil events, the time resolution obtainable for the Miocene record, which is low when compared to those of the Pliocene and Pleistocene (Olafsson, 1989, 1991; Rio et al., 1990a; Fornaciari et al, 1990,1993; Gartner, 1992). Furthermore, the purpose of these attempts was to test the nannofossil biostratigraphic reliability in the Miocene record, in terms of isochroneity of the bioevents and reproducibility over wide geographic areas. Here, we expand the existing database for middle and late Miocene nannofossils by showing quantitative distribution patterns of selected species in sections from the equatorial Indian and Pacific oceans, recovered during DSDPLeg 85 and ODP Legs 115,130, and 138 (Fig. 1 and Table 1). Quantitative distribution patterns are fundamental for evaluating reliability of the classic first occurrence (F0) and last occurrence (LO) events (Backman and Shackleton, 1983). In addition, detailed quantitative distribution patterns can provide supplementary events based on abundance fluctations (acme, absence— or "paracme"—intervals). These additional events can result in biostratigraphically useful data for regional and long-distance correlations in so far as they reflect regional paleoceanographic events and/or global evolutionary turnovers that do not result in extinctions. Quantitative distribution patterns, as well, are important for gaining insights into the paleoecology and biogeography of calcareous nannofossils and for overcoming many traditional problems of biostratigraphy, such as noises in the stratigraphic record (reworking, etc.)

1 Pisias, N.G., Mayer, L.A., Janecek, T.R., Palmer-Julson, A., and van Andel, T.H. (Eds.), 1995. Proc. ODP, Sci. Results, 138: College Station, TX (Ocean Drilling Program). 2 Universita di Parma, Istituto di Geologia, Italia. Present address: Universita degli Studi "G. D'Annuzio," Chieti-Facoltà di Scienze Matematiche, Fisiche e Naturali. 3 Universita di Padova, Dipartimento di Geologia, Paleontologia e Geofisica, Via Giotto 1, 35137 Padova, Italia. 4 Universita di Torino, Dipartimento di Scienze della Terra, Via Accademia delle Scienze 5, 10123 Torino, Italia.

and inconsistencies in data-collecting methodologies (see discussion in Backman and Shackleton, 1983; Rio et al., 1990b).

OBJECTIVES AND STRATEGY As stated above, our main objectives were to test the reliability of calcareous nannofossil biohorizons (biostratigraphic events) and to improve biostratigraphic resolution. To make our objectives and conclusions more clear, we found it necessary to state our concept of biostratigraphic reliability, a much discussed topic in the past (i.e., Gradstein, 1985; Hill and Thierstein, 1989; Rio et al., 1990b; Bralower et al.,1989). In our concept, a biohorizon is considered to be reliable when it is easily reproducible among the different researchers and can be consistently correlated among distant and/or different facies/ sections, maintaining its position relative to other biohorizons. This last property, recently named "ranking" by Gradstein (1985), is simply the scientific paradigm of biostratigraphy (the principle of faunal succession) on which William Smith at the end of the 18th century founded biostratigraphy (Prothero, 1990). Reproducibility of a biohorizon is function of (1) the clear taxonomy of the index species; (2) the mode of occurrence (subtle, abrupt, etc.) of the change in distribution pattern of the index species one chooses as biohorizon. Therefore, we will rank biostratigraphic reliability of an index species by evaluating the following: 1. Unambiguous taxonomy; 2. Mode of occurrence of the event by visual inspection of the abundance pattern ("morphology" of the event) of the index species; 3. Consistency of the relative position (ranking) with respect to other biohorizons in distant sections; and 4. Position of the events vs. the available chronomagnetostratigraphy.

MATERIAL AND METHODS The sections included in this study are located in Figure 1 and listed in Table 1, where pertinent references for location and background information on the 10 sites investigated and on the sites considered for comparison are reported. Except for DSDP Site 608,

479

I. RAFFI ET AL. Table 1. Summary of cores considered in this study.

Leg

Site/hole

DSDP 82

558

DSDP 82 DSDP 94 ODP 108 DSDP 85 DSDP 85 ODP 130 ODP 138 ODP 138 ODP 138 ODP 138 ODP 115 ODP 115 ODP 115 ODP 115

Location (lat, long)

Western North Atlantic Ocean (37°46.2'N, 37°20.61'W) 563 Western North Atlantic Ocean (33°38.53'N, 43°46.04'W) 608 Eastern North Altantic Ocean (42°50.20'N, 23°05.25'W) 667A Equatorial Atlantic Ocean (4°34.15'N, 21°54.68'W) 574 Central equatorial Pacific Ocean (4°12.52'N, 133°19.81'W) 575 Central equatorial Pacific Ocean (5°51.00'N, 135°02.16'W) 806 Western equatorial Pacific Ocean (0°19.11'N, 159°21.68'E) 844 Eastern equatorial Pacific Ocean (7°55.28'N, 9O°28.85'W) 845 Eastern equatorial Pacific Ocean (9°34.95'N, 94°35.45'W) 848 Eastern equatorial Pacific Ocean (2°59.63'S, 110°28.79'W) 853 Eastern equatorial Pacific Ocean (7°12.66'N, 109°45.08'W) 709 Western tropical Indian Ocean (3°54.72'S, 60°33.16'E) 710 Western tropical Indian Ocean (4°18.69'S, 60°48.76'E) 711 Western tropical Indian Ocean (2°44.46'S, 61°09.75'E) 714 Western tropical Indian Ocean (5°03.69'N, 73°46.98'E)

Water depth (m)

General lithology

References

3754

Nannofossil ooze

3786

3535

Khan et al. (1985) Foraminiferal nannofossil ooze and chalk Foraminifer and nannofossil ooze Clement and Robinson (1987) and chalk Mud-bearing nannofossil ooze

Bukry (1985); Miller et al. (1985); Parker (1985) Bukry (1985); Miller et al. (1985); Parker (1985) Takayama and Sato (1987); Olafsson (1991); Gartner (1992) Manivit (1989); Olafsson (1989)

4561

Calcareous ooze chalk

Pujos (1985); Olafsson (1989)

4536

Siliceous and nannofossil ooze

Pujos (1985); present study

3526

Khan et al. (1985)

2520.7 Foraminiferal nannofossil chalk

Fornaciari et al. (1993); present study Mayer, Pisias, Janecek, et al. (1992), Raffi and Flores (this volume); Schneider (this volume) present study Mayer, Pisias, Janecek, et al. (1992), Raffi and Flores (this volume); Schneider (this volume) present study Mayer, Pisias, Janecek, et al. (1992), Raffi and Flores (this volume); Schneider et al. (this volume) present study Mayer, Pisias, Janecek, et al. (1992)

3425.0 Clay-rich biogenic, silica-rich ooze + nannofossil ooze 3715.9 Diatom radiolarian, clay, and nannofossil ooze 3867.3 Foraminiferal nannofossil ooze 3727.2 Clayey nannofossil ooze 3046.9 Nannofossil ooze Schneider and Kent (1990) 3822.5 Clay-bearing nannofossil ooze Schneider and Kent (1990) 4438.7 Clay and clayey nannofossil ooze 2042.0 Foraminiferal nannofossil ooze

Figure 1. Location of DSDP and ODP Sites studied (*) and considered for reference (•).

all are located in low-latitude areas, but represent vastly different watermasses. The temporal extension of the various sections is summarized in Figure 2, where the availability of magnetostratigraphic records is evidenced. Light microscope techniques were used for examining smear slides, which were made directly from core samples, using standard methods. Quantitative data were collected according to three methods: 1. Counting the index species relative to the total assemblage. 2. Counting the index species relative to a prefixed number of taxonomically related forms (i.e., species of discoasterids relative to 200 discoasterids, etc.). 3. Counting the number of specimens of the index species per unit area of the slide.

480

Magnetics

Fornaciari et al. (1990); Rio et al. (1990); present study Backmanetal. (1990); Rio et al. (1990); present study Rio et al. (1990); present study Fornaciari et al. (1990); Rio et al. (1990); present study

Backman and Shackleton (1983) and Rio et al. (1990b) discussed these three methods at length, indicating their respective advantages as well as their limits of applicability. Method 1 was applied for evaluating abundances of Mynilitha convallis, and of all the index species reported at Site 806. Method 2 was applied for evaluating abundances of discoasterid species, Calcidiscus premacintyrei and C. macintyrei (>ll µm), Sphenolithus heteromorphus. Method 3 was applied for obtaining distribution patterns of ceratolithids and triquetrorhabdulid species and Coronocyclus nitescens. As regards Reticulofenestra pseudoumbilicus (>7 µm) and Cyclicargolithus floridanus, different counting methods were applied in the different sequences. R. pseudoumbilicus was counted vs. all the other nannofossils at Sites 575 (Fig. 3), 714 (Fig. 4), and 806 (Fig. 5), and vs. C.floridanusat Site 845 (Fig. 6). Except in this latter site, C. floridanus was counted vs. all the nannofossils at Sites 575 (Fig. 3) and 806 (Fig. 5) and per unit area at Site 714 (Fig. 4). Note that the different methods provide comparable biostratigraphic signals. As regards the magnetostratigraphic records reported in the investigated sequences, data are from Schneider and Kent (1990) for ODP Leg 115 sequences (Figs. 10 and 14), and from site chapters in Mayer, Pisias, and Janecek, et al. (1992) and Schneider (this volume) for ODP Leg 138 sequences (Figs. 6, 11-13, 15). We refer to the paleomagnetic time scale of Cande and Kent (1992), combined with the time scale developed for Leg 138 sites (Shackleton et al., this volume) (Fig. 17, back pocket).

REMARKS ON TAXONOMY Calcareous nannofossil species considered in this study are listed in the Appendix. Most of these are referenced in Perch-Nielsen (1985). Consistency in taxonomic concepts is a key factor in biostratigraphy, and the use of different taxonomic concepts (expecially when dealing with intermediate morphologies of evolving lineages) explains much of the differences in biostratigraphic ranges found in literature. We have made clear the adopted taxonomic concepts in previous papers

1980

Ch ro π

Polarity

(my)

Time

in

GPTS

Okada and Bukry

DISTRIBUTION PATTERNS OF CALCAREOUS NANNOFOSSILS

Equatorial Indian Ocean

709 5 6 7 -

C3n

C3r

5 —

710

Equatorial Pacitic Ocean

711

M M

CNlOc

North Atlantic Ocean

844

845

84F

85:

M

M

M

M

Λ Λ Λ

CNICa

C3Aπ

608

CN9b C3Ar C3Br C4n

g

M CN9a

•mi

C4r

9 10 11 -

•

\CNβb/~ C4AD

CNβa C4Ar

CN7

714

C5n CNβ

1 / ^

—, C5r

12 -

S06

575

CN5b

Illl

C5An

13 -

C5Ar CSAAn

CM 5a

C5ABπ C5ABr

14 -

C5ACa ^C5ACr

—

C5ADn

CN4

C5ADr

Figure 2. Position relative to integrated biomagnetostratigraphy of the investigated sections. "M" denotes sections with magnetostratigraphy. (Rio et al., 1990a; Fornaciari et al., 1990; Raffi and Flores, this volume), to which the reader is referred. When necessary, we will reiterate the adopted criteria in some of the index species determination, commenting on a single biohorizon.

RESULTS We established by different counting methods the distribution patterns of 29 middle and late Miocene index calcareous nannofossils, shown in Figures 3 to 16. Besides the distribution of index nannofossils, in some sequences we report also the distribution patterns of additional species, such as Discoaster musicus, D. aff. calcaris, and Discoaster sp. 2, which characterize the assemblages at different stratigraphic intervals. Specifically, D. aff. calcaris is a discoasterid observed both in tropical Indian and equatorial Pacific oceans within Zone CN5 (see Rio et al., 1990a, for description). Discoaster sp. 2 is a large discoasterid similar to Discoaster brouweri, consistently recorded within Subzone CN9b, in both low-latitude areas (Rio et al., 1990a; Raffi and Flores, this volume). Moreover, as regards discoasterids, we did not report quantitative distributions either of long-ranging species, which do not provide significative biostratigraphic signal (such as D. brouweri and D. variabilis), or of index species found as rare and scattered. Among this latter group, we included Discoaster braarudii, which does not result in a reliable biostratigraphic marker in the middle Miocene, as suggested by Gartner (1992). Both in tropical Indian Ocean sequences (ODP Leg 115) and equatorial Pacific Ocean sequences (ODP Leg 138), D. braarudii is rare, discontinuously distributed in Zones CN6 and CN7 (NN8 and NN9) and does not have any biostratigraphic utility. On the basis of the established distribution patterns, their long distance correlations, and their calibration to the available magnetostratigraphy (Fig. 17, back pocket), we discuss in the following sections the reliability of 37 biohorizons, shown in Table 2 and numbered

Table 2. Summary of biohorizons considered in this study. 37 36 35 34 33 32 31 30 29 28 27 26 25 24 23 22 21 20 19 18 17 16 15 14 13 12 11 10 9 7 6 5 4 3 2 1

Discoaster quinqueramus LO Amaurolithus amplificus LO Amaurolithus amplificus FO Amaurolithus primus F0 Discoaster surculus FCO Discoaster berggrenii F 0 Discoaster pentaradiatus FO Discoaster neorectus FO Discoaster loeblichii FO Discoaster bollii LO Mynilitha convallis F 0 Discoaster hamatus LO Discoaster neohamatus FO Discoaster prepentaradiatus F0 Catinaster calyculus LO Catinaster coalitus LO Discoaster hamatus FO Catinaster calyculus FO Coccolithus miopelagicus LO Discoaster bellus group F 0 Discoaster exilis LCO Discoaster calcaris F0 Catinaster coalitus F 0 Discoaster kugleri LO Discoaster kugleri LCO Discoaster bollii F 0 Discoaster kugleri FCO Discoaster kugleri FO Calcidiscus macintyrei FO Coronocyclus nitescens LO Triquetrorhabdulus serratus LO Calcidiscus premacintyrei LCO Triquetrorhabdulus rugosus FCO Discoaster signus LO Cyclicargolithus floridanus LCO Reticulofenestra pseudoumbilicus F 0 Sphenolithus heteromorphus LO

Notes: LO = last occurrence; FO = first ocurrance; LCO = last common occurrence.

481

I. RAFFI ET AL.

N/mm

30 0

N/mm

90 0

N/mm

90 0

90

0

Figure 3. Abundance patterns of middle Miocene selected calcareous nannofossils atHole 575B. (mbsf) = meters below seafloor. N/mm2 = number of specimens per square millimeter; FO = first occurrence; LO = last occurrence; FCO = first common and/or continuous occurrence; LCO = last common and/or continuous occurrence.

Table 3. Position of calcareous nannofossil events at Hole 575B. Event D. hamatus FO C. miopelagicus LO C. nitescens LO T. serratus LO C. macintyrei FO C. premacintyrei LCO T. rugosus FCO C.floridanus LCO S. heteromorphus LO R. pseudoumbilicus F0

Core, section (cm)

Depth (mbsf)

5H-3, 115/5H-2.28 5H-3, 115/5H-2.28 7H-2, 49/7H-1.95 7H-2, 49/7H-l,95 7H-6, 52/7H-5, 5 7H-6, 52/7H-5, 5 8H-3, 31/8H-2, 90 9H-2.93/9H-1.32 9H-5, 137/9H-4, 92 10H-3, 1O8/1OH-2, 47

43.15-40.78 43.15-^0.78 57.39-56.35 57.39-56.35 63.42-61.45 63.42-61.45 68.01-67.1 76.23-74.12 81.17-79.22 85.48-83.87

Notes: F0 = first occurrence; LO = last occurrence; FCO = first common and continuous occurrence; LCO = last common and continuous occurrence. in stratigraphic order. The stratigraphic position in the single sections of these biohorizons is summarized in Tables 3 to 13.

LO of Sphenolithus heteromorphus (1) The LO of the easily identified Sphenolithus heteromorphus (definition of the top of Zones CN4 and NN5) appears as one of the most easily determined and correlatable event in the investigated sections (Figs. 3, 5, and 6), as is generally acknowledged in the literature (i.e., Olafsson 1989,1991). Problems with the LO of S. heteromorphus arise with calibration to the GPTS. In fact, Berggren et al. (1985) associated

Table 4. Position of calcareous nannofossil events at Hole 714A. Event M. convallis FO D. hamatus LO D. pentaradiatus FO D. bollii LO D. prepentaradiatus FO C. coalitus LO D. neohamatus FO D. hamatus F 0 D. bellus group F0 C. miopelagicus LO C. calyculus F 0 D. exilis LCO D. kugleri LO D. calcaris FO C. coalitus FO D. bollii FO D. kugleri FO C. nitescens LO T. serratus LO R. pseudoumbilicus FO S. heteromorphus LO T. rugosus FCO C.floridanus LCO Notes: As specified in Table 3.

Core, section (cm)

Depth (mbsf)

4H-7, 140/4H-7, 75 5H-1.30/4H-7, 140 5H-1, 140/5H-1.30 5H-4, 30/5H-3, 140 6H-1.30/5H-7, 30 7H-1.30/6H-7-30 7H-2, 140/7H-2, 30 7H-4, 30/7H-4, 10 7H-4, 30/7H-4, 10 7H-4, 75/7H-6, 10 Not reliable 7H-7, 30/7H-6, 140 8H-1.75/7H-7, 30 8H-1, 140/7H-7, 30 8H-1, 140/8H-l,75 8H-4, 75/8H-3, 140 Not reliable 10H-1.30/9H-5, 30 10H-1.30/9H-5, 30 10H-6, 75/1OH-5, 130 10H-6, 75/10h-5, 130 10H-6, 140/10H-6, 30 10H-6, 140/10H-6, 30

31.7-31.05 32.0-31.7 33.1-32.0 36.5-36.1 41.7-41.0 51.3-50.7 53.9-52.8 55.8-55.6 55.8-55.6 58.6-56.25 60.30-59.9 61.35-60.3 62.0-60.3 62.0-61.35 65.9-65.0 80.1-76.5 80.1-76.5 88.05-87.1 88.05-87.1 88.7-87.6 88.7-87.6


Φ Q-.2

al•

O ü

N/mm

N/mm

60

0

0

20

60

N/mm

N/mm

20

0

0

20

70-

9H

LO 80-

-

10H

F0 FCO

90-

-

1 1H

12H

Figure 4. Abundance patterns of middle Miocene selected calcareous nannofossils at Hole 714A. (mbsf) meters below seafloor. Notation as specified in Figure 3.

Figure 5. Abundance patterns of middle Miocene selected calcareous nannofossils at Hole 806B. (mbsf) = meters below seafloor. Notation as specified in Figure 3.

483

I. RAFFI ET AL.

Figure 6. Abundance patterns of middle Miocene selected calcareous nannofossils at Site 845. (mcd) = meters composite depth. Notation as specified in Figure 3. C.c. = Catinaster coalitus; D.k. = Discoaster kugleri; S.h. = Sphenolithus heteromorphus. Magnetostratigraphy from Schneider (this volume).

Table 5. Position of calcareous nannofossil events at Hole 806B. Event D. kugleri FO C. nitescens LO C. macintyrei F0 T. serratus LO C. premacintyrei LCO T. rugosus FCO C.floridanus LCO S. heteromorphus LO R. pseudoumbilicus F0

Core, section (cm) Not detected 45X-6, 60/45X-5, 50 Not detected 45X-6,60/46X-l,50 46X-2,60/46X-l,50 48X-3, 50/49X-2, 60 50X-l,70/49X-7,30 51X-l,70/50X-6,60 51X-6.60/51X-7, 30

Depth (mbsf) 425.0-423.4 427.0-425.0 428.6-427.0 459.0-457.6 464.3^64.2 474.0-471.7 482.6-481.4

Notes: As specified in Table 3.

this important event with Chron 5ADn, considering the results of Miller et al. (1985) at DSDP Sites 558 and 563, whereas Backman et al. (1990) and Olafsson (1991) associated it to Chron 5ABr on the basis of results from DSDP Site 608. About 1 m.y. of diachroneity is inferred when accepting these data. However, note that recently, Miller et al. (1991) correlated, via oxygen isotope stratigraphy, Sites 563 and 608, and reinterpreted the magnetostratigraphic record of the former succession (compare Fig. 1 in Miller et al., 1985, with Fig. 4 in Miller et al., 1991). At both sites, S. heteromorphus LO is almost coincident

with a prominent stable oxygen isotope minimum in benthic foraminifers (Mi3), which in the new interpretation of Miller et al. (1991), is associated in both sites with Chron 5ABr. A similar calibration can be inferred also at Site 521 (Hsü et al., 1984) and at Indian Ocean Site 710 (Rio et al., 1990a), where the LO of S. heteromorphus clearly postdates Chron 5AD time. No reliable magnetostratigraphy is available in the sequences of Leg 138 sediments in the interval around the LO of S. heteromorphus. However, extrapolating sediment accumulation rates at Site 845, an age corresponding to the top Chron 5 ACn—base Chron 5 ABr time— is obtained (Raffi and Flores, this volume). We consider the S. heteromorphus LO an excellent biostratigraphic event, which has probably a high chronostratigraphic correlation potential between low- and mid-latitude areas (Fig. 17, back pocket).

FO of Reticulofenestra pseudoumbilicus (2) For distinguishing Reticulofenestra pseudoumbilicus, we considered reticulofenestrids larger than 7 µm, following the taxonomic concepts of Raffi and Rio (1979) and Backman and Shackleton (1983). These reticulofenestrids provide a biostratigraphically and biochronologically reliable event in the Pliocene, when they became extinct in the upper part of Chron 2Ar (late Gilbert) (Backman and Shackleton, 1983; Rio et al. 1990a, 1990b; Raffi and Flores, this


-C js. Q-

_Q

oi,