18. Preliminary Results of High-resolution Pollen Analyses of Selected Pleistocene Intervals from Holes 798a, 798b, and 798c

Results of preliminary pollen analyses of 280 samples from selected intervals of cores from Hole 798 on the Oki Ridge in the Sea of Japan (Sections 798A-1H-1 through 798A-3H-3,798B-1H, 798B-13H1 through 798B-15H-8, and 798C-1H-1 through 798C-1H-3) indicate that (1) diagnostic pollen types of the vegetation of Japan are abundant and well-preserved in sediments deposited over the last 1.4 m.y.; (2) the concentration, preservation, and composition of pollen in the rapidly deposited sediments of Site 798 provide a basis for high-resolution (on the order of ~500-year intervals) pollen stratigraphy in Pliocene-Pleistocene Japan Sea sediments; (3) the pollen stratigraphy of sediments from Site 798 can be directly correlated with pollen stratigraphies of correlative ages onshore; and (4) the development of continuous, directly correlative Pliocene-Pleistocene terrestrial/marine paleoclimatic records from sediments deposited at Site 798 will provide a continuous high-resolution record of northeast Asian marine and terrestrial paleoenvironmental changes for the past -2.4 Ma. INTRODUCTION A prime objective at Site 798 was to develop a Neogene paleoceanographic reference section. Taken close to the main island of Japan, the retrieved section, in which sedimentation rates average 120 m/m.y., apparently provided a unique opportunity to develop a high-resolution terrestrial paleoenvironmental record (Koroneva, 1961; Morley et al., 1986). My initial objective was to evaluate the suitability of pollen in sediments from Hole 798A for detailed paleoecological and biostratigraphic analyses. Because the preliminary pollen analyses suggested that several meters of the topmost sediment had not been retrieved in Hole 798A and the Initial Reports (Ingle, Suyehiro, von Breymann, et al., 1990) showed that Holes 798B and 798C had retrieved the mud/water interface, the uppermost sediment from Holes 798B and 798C was also analyzed for pollen in order to obtain a more complete Quaternary overview. Study Area Site 798, in the southeastern Sea of Japan (37.038°N, 134.80°E, 900 mbsf) is on the Yamato Rise, a fault-bounded topographic high located -100 km west of Honshu, Japan (Fig. 1). The proximity of this site to Japan suggests, and empirical evidence confirms, that the primary source of pollen on the Oki Ridge is the vegetation of Japan (Heusser, 1977, 1988; Heusser and Morley, 1985; Sarro, 1991). On the back side of Japan, potential natural montane vegetation, which varies with latitude and altitude, consists of evergreen and deciduous broadleaf and coniferous forests (Numata, 1974; Heusser, this volume). Characteristic dominants of the warm-temperate forests include evergreen members of the Fagaceae, such as Quercus and Castanopsis, along with conifers such as Cryptomeria japonica, Podocarpus, and the secondary species Pinus densiflora. Temperate, deciduous broadleaf forests are characterized by Quercus/Fagus associations in which Pinus and Cryptomeria are present. Conifers (Picea, Abies and Tsuga) along with Betula and Alnus are associated with subalpine elevations (1500-2000 m above sea level) (Ohwi, 1984). Pisciotto, K. A., Ingle, J. C, Jr., von Breymann, M. T., Barron, J., et al, 1992. Proc. ODP, Sci. Results, 127/128, Pt.l: College Station, TX (Ocean Drilling Program). 2 Lamont-Doherty Geological Observatory of Columbia University, Palisades, NY 10964, U.S.A. Material and Methods All three holes at Site 798 (Fig. 1, Table 1) were sampled for pollen. From the silty and siliceous silty clays of the upper 27 mbsf of Hole 798A (Samples 798A-1H-1, 1-3 cm, through 798A-3H-3, 76-78 cm), 177 samples for joint pollen/stable isotope high-resolution paleoclimatic study were taken at consecutive ~ 10 cm intervals. Forty samples were processed from the uppermost sediment from Holes 798B and 798C. Samples from the upper 125 cm of Core 798B-1H were taken from contiguous intervals of whole-round-cores (Cragg, this volume). The sampling intervals from Hole 798C are more variable and about six times as large as the sampling interval in Core 798B-1H. Samples 798B-13H-1,0-30 cm, through 798B-15H-8, 30-60 cm, are splits of 63 contiguous "scrape" samples that integrate -30 cm in depth (P. DeMenocal, pers. corara.). Sample preparation procedures and methodology used in pollen identification are described elsewhere (Heusser, this volume). Pollen data, selected from the pollen data sets on file in the ODP databank, are here presented as percentages of the preliminary pollen counts (>IOO pollen grains/sample, and as concentration (pollen/grams dry weight sediment = gdws). Because of the preliminary nature of the pollen data, pollen influx was not determined nor were the pollen diagrams divided into pollen assemblage zones. Chronostratigraphy Unless otherwise stated, age models used in this paper are based on sedimentation rates of -12 cm/k.y. (Table 2). The age of the uppermost sediment recovered in Holes 798B and 798C was presumed to be Recent because the mud line was apparently retrieved (Ingle, Suyehiro, von Breymann, et al., 1990). Estimates of the depth of uppermost sediment in Hole 798 A ranged from 1 to 4 m below the mud/water interface (Ingle, Suyehiro, von Breymann, et al., 1990). The chronology of samples from the upper 30 mbsf of Hole 798A is further complicated by the nonretrieval of sediment between 13.5 and 18.5 mbsf (Ingle, Suyehiro, von Breymann, et al., 1990). Comparison of the pollen stratigraphy of Cryptomeria from Section 798A-1H-1 with that of nearby piston cores V28-265 and RC14-99 (Fig. 2, Table 2) suggests that the age of the uppermost sediment in Hole 798 A may be as old as 65 k.y. (Morley et al., 1986). Comparison of the pollen stratigraphy of the uppermost sediment deposited at Holes 798A, 798B, and 798C with pollen stratigraphy from piston core V28-265 indicates that the uppermost sediment retrieved in Hole 798B and 798C-1H is Holocene (Morley et al., 1986; Sarro, 1991).