Biostratigraphic analyses can provide ages for the fossil contents of sediments that should represent the age of original deposition. For sections interpreted as having undergone secondary transport in the form of mass movement, biostratigraphy can only provide information concerning the minimum age of the sediments immediately underlying the redeposited interval. Thus, the age of the underlying sediment indicates only a maximum age for the redistribution event itself. The early Eocene age of the Radiolaria in Subintervals B3–B4 and the more speculative similar ages for Intervals C–D indicate only a maximum age for a stratigraphic section assumed to have arrived at this location by mass wasting at a much younger age. Biostratigraphic analyses of sediments above and below a turbidite can restrict the time frame of the transport event. Despite the absence of microfossils in our smear slides from lithologic Units 3 and 4, we investigated six samples from this section with the objective of finding enough Radiolaria specimens to further constrain the ages of the turbidites in lithologic Unit 2. However, Radiolaria younger than Eocene were not observed in these samples, with the exception of a trace occurrence of Euchitonia furcata in the sample at 8.75 mbsf. The Neogene Radiolaria fauna in Subintervals B1–B2 provide a minimum age of 1.61–1.8 Ma for the turbidites below this interval and a maximum age of 2.0 Ma for the turbidites of Interval A.
The occurrences of Radiolaria observed in the sediments in Hole 1223A are consistent with their presence there as a result of the Nuuanu landslide. Abundant and well-preserved Eocene Radiolaria have an almost global distribution in pelagic sediments (Baldauf and Barron, 1990), and the presence of this Eocene fauna in sediments derived from the sedimentary apron of Koolau Volcano requires no special explanation; however, the presence of the Neogene fauna at this location in the central gyre of the North Pacific is not easily explained. Siliceous microfossil dissolution proceeds to essential completion in the low-productivity regions of the central gyres of the modern oceans. Possibly, the North Pacific central gyre experienced a brief surge in productivity during the late Pliocene that resulted in elevated delivery rates of siliceous skeletal debris arriving on the seafloor and consequently improved conditions for microfossil preservation. Alternatively, the presence of seafloor sediments containing high concentrations of volcaniclastic components as a result of the Nuuanu landslide may have produced environments where silica dissolution was sufficiently depressed to allow partial preservation of siliceous skeletal debris. This latter explanation, however, is not consistent with the distribution of Radiolaria in Hole 1223A immediately above Subintervals B3–B4, where opal-A preservation is very poor.
Stephens, Kasahara, Acton, et al. (2003) postulated that the two vitric tuffs of lithologic Units 5 and 11 may have resulted from a very large eruption of primitive tholeiite that occurred when a deep magma reservoir was breached during failure of the northeast flank of Koolau Volcano, thus resulting in the giant Nuuanu debris avalanche. The volcaniclastic sand resulting from this eruption and the sediments entrained with it are believed to have been swept north at tremendous speed as a submarine debris flow. These authors suggest that the heat retained within the debris flow caused alteration and induration of the two vitric tuffs. This heat source may have also been responsible for the opal-A to opal-CT alteration of Radiolaria in the claystones and siltstones below the tuffs (Interval D). Garcia et al. (2006) present evidence against the hot debris flow model and argue that the tuffs are more correctly described as sandstones; however, we observe that the degree of opal-CT recrystallization in Interval D appears to increase with depth. Consequently, the heat source responsible for induration of the sediments below Core 200-1223A-2H and opal-CT recrystallization of Interval D may be located in the sediment column below 41 mbsf at Site 1223.
Similarities between the Eocene Radiolaria assemblages preserved in Hole 1223A sediments suggest that the entire section was derived from the same source location. The mixed ages of the Radiolaria assemblage above Subinterval B3 may indicate that the source of these younger turbidites was shifting to a different location on Koolau, where late Pliocene sediments were available for transport and redeposition.
The improved state of preservation of the Radiolaria in Interval B is in marked contrast to the highly fragmented condition of the skeletal lattice observed in Interval A. We commented above on the unusual nature of the sediment at 6.7 mbsf, where abundant and well-preserved Radiolaria are associated with a very high concentration of volcaniclastic grains. Whereas the condition of the Radiolaria in Interval A is consistent with turbidite transport over long distances, the good to very good preservation of specimens in Subintervals B3–B4 suggests a less abrasive depositional process. Perhaps Subintervals B3–B4 arrived at Site 1223 as an intact unit within the Koolau debris avalanche. Alternatively, Subintervals B3–B4 may have been emplaced as a single, local slump derived from the one of the numerous blocks and seamounts of the Nuuanu-Wailau debris field. Intermittently abundant occurrences of tabulate crystals suspected of being opal-CT blades in the sediments of Intervals A and B may be further evidence of local weathering and redeposition of sedimentary constituents derived from outcrops of Interval D claystones and siltstones.
Mixed Radiolaria assemblages composed of faunas with ages similar to those of Hole 1223A occur elsewhere in the central North Pacific. At Deep Sea Drilling Project (DSDP) Site 40 (~1700 km east-southeast of Site 1223), the upper 10 m of the stratigraphic section consists of zeolitic red clay containing a Radiolaria assemblage predominantly comprising Eocene taxa with minor late Neogene constituents. This interval is immediately underlain by 143 m of Eocene Radiolaria ooze (McManus et al., 1970). Unfortunately, a detailed biostratigraphic analysis of the upper portion of this hole is not available. Benthic burrowers may account for the fossil admixture at Site 40, but the vast preponderance of the Eocene admixed constituents is troubling. At other localities where burrowing by benthic organisms is obvious, the older and deeper fauna that has been reworked up through the sediment column represents only a minor percentage of the total fossil assemblage. Inexplicably, the upper sediments at Site 40 contain about 95% Eocene taxa admixed with minor concentrations of Pleistocene species. Sedimentary reworking by burrowers is an even less probable explanation for the mixed assemblage present in Hole 1223A, where the immediately underlying sediments are barren.
Important similarities and distinctions exist between the distributions of Radiolaria in the section recovered at Site 1223 and those of ODP Leg 136 Sites 842 and 843, which are located 320 km west of the island of Hawaii. The stratigraphic sections at Sites 842 and 1223 both include intervals ~1 m thick containing mixed late Neogene and Eocene Radiolaria within stacked turbidites. In the Leg 136 recovery, these mixed faunas are restricted to three ash layers in the 7.5–9.6 mbsf interval of Hole 842A (Garcia and Hull, 1994). These ash beds are interspersed in a succession of brown clay and clayey silts containing common to abundant, exclusively Neogene Radiolaria variously dated as Quaternary (Hull, 1993), Pleistocene (Garcia and Hull, 1994), or assigned to the lower Pleistocene Amphirhopalum ypsilon Zone (Hull, 1993). More precise age determinations for this assemblage is precluded by the absence of critical index taxa. The magnetostratigraphic record of Holes 842A and 842B suggests that the ash beds were deposited during early Magnetochron C1r2r (Helsley, 1993; Garcia and Hull, 1994). This succession at Site 842, containing an undiluted Neogene Radiolaria assemblage, is replaced by the largely barren turbidites of Interval A in Hole 1223A. Moreover, admixture of Neogene and Eocene faunas is not restricted to volcaniclastic sediments in Interval B of Hole 1223A, and the recovery from Holes 842 and 843 lack beds containing purely Eocene Radiolaria, as represented in Subinterval B2. The black, indurated claystones cemented with opal-CT of Cores 136-842B-3H and 4H (Shipboard Scientific Party, 1992) may be correlative with lithologic Units 5 and 11 of Site 1223.