RESULTS AND DISCUSSION

Carbonate contents show an irregular distribution with depth in Holes 1150A and 1151A (Figs. F2, F3), ranging from 25 wt% to values below the detection limit of 0.01 wt% (Tables T1, T2). Carbonate contents in ocean sediments depend on a number of factors including (1) productivity of carbonate shell-bearing organisms, (2) variation of the carbonate compensation depth (CCD), (3) dilution by detrital material, and (4) postdepositional (diagenetic) processes (Libes, 1992). The presence of well-preserved calcareous tests of marine organisms throughout the studied sedimentary column at Sites 1150 and 1151 (Sacks, Suyehiro, Acton, et al., 2000) strongly suggests that sedimentation occurred above the CCD. Currently, the North Pacific CCD occurs at a depth of ~3000 m, although it rises to a depth of ~2400 m in coastal and highly productive areas of the North Pacific Ocean (Rea et al., 1995). The water depth of Sites 1150 and 1151 are 2681 m and 2182 m, respectively. It is then possible that variations in carbonate content in the studied sediments are associated with fluctuations in the CCD, given that the modern CCD occurs near the water depth of these sites. However, the reconstructed North Pacific CCD indicates that the modern depth has been the shallowest during the Cenozoic (Rea et al., 1995). Consequently, it appears unlikely that the ~1300- to 1600-m rise of the CCD in the North Pacific Ocean during the last 25 m.y. could have significantly affected carbonate contents in sediments recovered from Sites 1150 and 1151 because these sites were most likely above paleo-CCD.

One of the most important diagenetic processes affecting carbonate dissolution is microbial degradation of organic matter (OM) in ocean sediments (Berger et al., 1982). Carbon dioxide formed during the microbial oxidation of OM shifts the carbonate equilibria in pore water, resulting in the dissolution of carbonate minerals (Berger, 1970; Hales and Emerson, 1996). The rate of interstitial carbon dioxide production depends on the level of oxycity in the sediments and on the type of OM being delivered to the ocean bottom (Calvert, 1987). In general, algal OM is more easily degradable than terrestrial OM (Opsahl and Brenner, 1995), potentially resulting in faster rates of carbon dioxide production. Although OM deposited at Sites 1150 and 1151 is mainly marine in origin (see below) and consequently is likely to be readily oxidized, dissolution caused by this oxidation was probably minimum as indicated by the good preservation of foraminiferal tests in the sediments (Sacks, Suyehiro, Acton, et al., 2000). However, some carbonates at Sites 1150 and 1151 correspond to dolomite (Sacks, Suyehiro, Acton, et al., 2000), strongly suggesting a diagenetic origin.

Mass accumulation rates (MARs) of carbonates are a better representation of the delivery of carbonates to the ocean floor because they account for dilution effects caused by changing sedimentation rates. The age model employed to estimate MAR for carbonates at Leg 186 sites is based on a combination of biostratigraphic events and magnetic polarity (Sacks, Suyehiro, Acton, et al., this volume). Estimations of MAR for carbonates vary from values as high as 30 g/m²/yr to almost negligible values (Figs. F4, F5). Notably, high MAR values occur in sediments accumulated between 5 and 7 Ma. Although diagenetic formation of carbonates can potentially augment MAR, increased delivery of calcareous tests from planktonic organisms seems a more plausible cause for the observed high MAR values between 5 and 7 Ma. An independent evaluation of the possibility that increased productivity of planktonic organisms produced the observed high MAR values between 5 and 7 Ma at Sites 1150 and 1151 can be achieved by studying accumulation of organic carbon matter, which is another proxy for primary sea-surface productivity.

Organic matter contents in sediments recovered at Sites 1150 and 1151 show relatively moderate values, ranging from ~0.5 to 1.5 wt%. A decreasing trend in organic carbon abundance occurs with depth at both sites (Figs. F2, F3). Although organic carbon contents of hemipelagic sediments are generally related to productivity, changes in sedimentation rates can strongly affect organic carbon contents by diluting accumulating OM. Mass accumulation rates of organic carbon, however, account for changing sedimentation rates, thereby better representing the delivery of OM to the sediments. MAR for organic carbon in Holes 1150A and 1151A ranges from ~ 0.2 to 3 g C/m²/yr (Figs. F4, F5). Relatively high values occurred in sediments deposited between 5 and 6.8 Ma and between 0.5 Ma and recent in Hole 1150A (Fig. F4). Increased delivery of OM between 5 and 7 Ma is also evident in Hole 1151A (Fig. F5). However, the significant increase in OM accumulation at Site 1150 during the last 0.5 m.y. is unresolved in the record of Hole 1151A.

Increased delivery of organic carbon between 5 and 7 Ma can be explained by (1) enhanced input of terrigenous OM, (2) increased productivity of marine organisms, (3) increased accumulation rates that enhanced the preservation of OM by reducing the exposure time to oxic conditions, or (4) formation or enhancement of anoxic conditions at the sediment/water interface (Emerson and Hedges, 1988). An evaluation of the source of OM and the sedimentary conditions during OM accumulation assists with the examination of these possibilities.

Organic carbon/nitrogen (C/N) ratios can be used to assess OM sources because terrestrial OM typically shows C/N ratios >15, whereas marine algal OM exhibits C/N ratios <10 (Tyson, 1995). This difference in C/N ratios is due to the abundant presence of N-depleted compounds in vascular plants, such as lignin and cellulose. C/N ratios in sediments at Sites 1150 and 1151 range from 2.5 to 12, indicating a predominantly marine source for the accumulated OM (Figs. F2, F3). Original C/N ratios, however, can be affected by diagenetic factors. For example, a decreasing trend in C/N ratios is observed at Sites 1150 and 1151. This trend is probably the result of degradation of organic matter as suggested by a correlative decreasing trend in organic carbon content with depth (Figs. F2, F3). Another diagenetic process affecting C/N ratios is the incorporation of ammonia produced during OM degradation in the interlayer positions of clay minerals. This process decreases C/N ratios of bulk nitrogen determinations (Müller, 1977), and it can be identified when C/N ratios are lower than those existing in marine organisms. Whereas photosynthetic algae typically exhibit C/N ratios of ~4 (Redfield et al., 1982), some intervals at Sites 1150 and 1151 exhibit C/N ratios of <4 (Figs. F2, F3), thereby indicating the diagenetic incorporation of ammonia in clay minerals.

Although C/N ratios at Leg 186 sites fall within those reported for marine algae, some variability in these ratios exists in sediments from Holes 1150A and 1151A, probably as a result of different degrees of OM degradation. Depending on pore water conditions and the nature of OM, it is possible that some intervals have experienced a larger degree of microbial degradation that produced various degrees of OM alteration. For example, sedimentation rates can potentially control the degree of OM degradation by determining the residence time of OM in oxic zones. At high sedimentation rates, settling OM is rapidly buried and removed from the water/sediment interface, where a significant degree of degradation occurs (Müller and Suess, 1979). No apparent correlation exists, however, between sedimentation rates and C/N ratios at Sites 1150 and 1151 (Figs. F3, F4). Alternatively, changes in the quantity of terrestrial OM delivered to the ocean floor can also account for the observed C/N variability at Sites 1150 and 1151, due to the presumed resistance of terrestrial OM to microbial degradation.

Another parameter used to distinguish terrestrial and marine OM sources is the carbon isotopic composition of OM (¹³C_org). Whereas terrestrial vascular plants use atmospheric CO₂ as a carbon source and result in ¹³C_org values of about -27 ± 3, algae assimilate dissolved CO₂ or bicarbonate and result in ¹³C_org values of about -21 ± 2 (O'Leary, 1988). This difference in ¹³C_org values helps to evaluate the relative contribution of terrestrial and marine OM (e.g., Prahl et al., 1994). Values of ¹³C_org in Hole 1150A sediments range from -23.4 to -21.3 (Fig. F2), suggesting a contribution of predominantly marine OM to the sediments. This suggestion is in agreement with the marine origin for the accumulated OM inferred from C/N ratios. Values of ¹³C_org show a decreasing trend with depth at Site 1150. This trend parallels that of organic carbon contents and C/N ratios, thereby indicating that microbial degradation of sedimentary OM produced a preferential loss of compounds enriched in ¹³C.

Superimposed on the decreasing trend, ¹³C_org values show a positive excursion between 3 and 4 Ma, followed by a negative shift (Fig. F2). These excursions may be the result of changes in the dissolved CO₂ that produced variations in marine algae ¹³C_org values (Rau et al., 1992; Hayes, 1993). In today's oceans, there is a latitudinal gradient in the isotopic composition of marine algae, showing relatively low values near the poles and relatively high values in the tropical oceans. This isotopic gradient appears to obey the latitudinal variation in sea-surface temperatures that control CO₂ solubility (Rau et al.,1989). Thus, variations in sea-surface temperatures during the last 4 m.y. could explain the observed ¹³C_org trend at Site 1150. In addition to temperature, productivity influences the isotopic composition of marine algae. At oligotrophic conditions, algae can maximize the preferential discrimination of ¹³C during photosynthesis, resulting in relatively low ¹³C_org values. In contrast, high productivity typical of upwelling zones causes algae to show relatively high ¹³C_org values (e.g., Pankost et al., 1999). The possibility that changes in productivity produced the observed excursions in ¹³C_org values at Leg 186 sites seems unlikely, as indicated by the lack of correlation between ¹³C_org values and elevated C_org contents and organic carbon accumulation rates (Fig. F2). In fact, the presence of diatomaceous sediments at Sites 1150 and 1151 suggests relatively high primary productivity. Consequently, changes in carbon isotope ratios are most likely due to changes in solubility of CO₂ controlled by temperature fluctuations.

Results at Leg 186 sites indicate that primary productivity has been relatively high, as manifested by the presence of diatomaceous sediments throughout the studied cores and by relatively high organic carbon and carbonate contents. There was, however, an even higher productivity event between 7 and 5.3 Ma, as indicated by high MAR values for carbonates and organic carbon. Similar results of enhanced productivity for this time period have been obtained for a number of records in the Indian Ocean and the Pacific Ocean (e.g., Peterson et al., 1992; Dickens and Owen, 1996; Seisser, 1995). Because the residence time of limiting marine nutrients (nitrate and phosphate) is shorter than the 2-to 4-m.y.-long algal bloom interpreted for the Pacific and Indian Oceans, a mechanism is needed to explain both augmented surface productivity and the sequestering of nutrients in sediments. The invoked mechanisms include an external (continental) delivery of nutrients and an internal (oceanic) supply of nutrients from the deep ocean. Delaney and Filippelli (1994) and Rea et al. (1995), among others, argue that this period of enhanced productivity was the result of higher input of nutrients resulting from weathering of the Himalayan Mountains. Under this scenario, the Himalayas reached a height sufficient to trap storm tracks from the Indian Ocean at ~8 Ma, producing an intensification of the Asian monsoon and the consequent net increased transfer of nutrients from land to oceans (Filippelli, 1997). An additional supply of nutrients to the world's oceans could come from the South American continent as a result of the main uplift of the Andes Mountains, which also occurred in the late Miocene (Gregory-Wodzicki, 2000).

Although enhanced weathering explains the observed increase in MAR, another possible scenario involves increased nutrient supply that resulted from a more vigorous oceanic circulation, which transferred nutrients to the Pacific and Indian Oceans from other oceanic localities (Dickens and Owen, 1996; Hermoyian and Owen, 2001). The implications for each mechanism are different. Enhanced continental delivery of nutrients to the oceans promoting elevated productivity would result in a net drawdown of atmospheric carbon dioxide as marine organic carbon and carbonate burial increased. In contrast, a redistribution of nutrients caused by an intensification of ocean circulation would result in enhanced productivity in some areas but in decreased productivity in others, with no detectable change in atmospheric CO₂ concentrations. Although results from Sites 1150 and 1151 cannot resolve any of these mechanisms, they do indicate that the productivity event of the late Miocene and early Pliocene also occurred in the northwestern Pacific, and it is consistent with the suggestion that this productivity event resulted from the establishment of the modern deep-water circulation that brings nutrients to the Oyashio Current (Rea et al., 1995).