18. STRATIGRAPHIC RESOLUTION OF LEG 85 DRILL SITES: AN INITIAL ANALYSIS 1 Nicklas G. Pisias, Oregon State University John A. Barron, U.S. Geological Survey, Menlo Park Catherine A. Nigrini and Dean A. Dunn, University of Southern Mississippi 2 ABSTRACT Biostratigraphic datums determined from the study of radiolarian and diatom microfossils are combined with varia- tions in calcium carbonate content to provide an estimate of stratigraphic resolution for the Leg 85 drill sites. These da- ta sets are used to correlate sites, utilizing techniques which consider the sampling resolution of the primary data. The graphic correlation procedure of Shaw (1964) provides an objective means for estimating the line of correlation between sites. The two microfossil groups are used independently to define two lines of correlation. The inverse mapping tech- nique of Martinson et al. (1982) is used to correlate sites from calcium carbonate data. Comparison of the lines of cor- relation determined from these three data sets suggests that stratigraphic resolution is on the order of 125,000 to 380,000 yrs. However, these limits of resolution are of the same order as the sampling density for the biostratigraphic and carbonate analyses. This suggests that a higher degree of stratigraphic resolution is possible at these important hy- draulically piston-cored sites. INTRODUCTION One of the primary objectives of Leg 85 and other hydraulic-piston-coring (HPC) cruises aboard Glomar Challenger was to provide high-quality, continuous sedi- ment sections for "high-resolution" stratigraphic and pa- leoceanographic studies. The ability to carry out such detailed studies depends in part on the quality of the stratigraphic framework into which sediment sections can be placed. On Leg 85, five sites were drilled in the eastern and central equatorial Pacific. Coring at four sites, 572, 573, 574, and 575, recovered at least 150 m of sediment using the hydraulic piston corer. By double piston coring most of the upper portion of the sediment column, continu- ous high-quality sediment sections were recovered. Strati- graphic correlation of these sites is based primarily on biostratigraphic datums. Magnetic stratigraphy is avail- able only in the intervals at Sites 573, 574, and 575 where sediment accumulation rate was lower; thus, it provides only limited stratigraphic resolution (Weinreich and Theyer, this volume). Barron et al. (this volume) have summarized the paleontological and magnetic stra- tigraphy available for sites drilled during Leg 85. In this chapter we use a different approach to address the ques- tion of stratigraphic resolution. We try to define the pre- cision of correlation attainable for Leg 85 sites and pre- sent a strategy for defining high-resolution stratigraphies for marine sediment sections using biostratigraphic and carbonate stratigraphies. 1 Mayer, L., Theyer, F., et al., Ml. Repls. DSDP, 85: Washington (US. Govt. Printing Office). 2 Addresses: (Pisias) College of Oceanography, Oregon State Univ., Corvallis, OR 97331; (Barron) U.S. Geological Survey, Menlo Park, CA 94205; (Nigrini) 510 Papyrus Drive, La Habra Heights, CA; (Dunn) Department of Geological Science, Univ. So. Miss., Hatties- burg, MS 39406. METHODS Our strategy is to correlate all sites in detail, using diatoms and ra- diolarians, and to correlate using the characteristics of the calcium carbonate records of these pelagic sediment sections. The radiolarian and diatom data are used independently to provide two estimates of the correlation between sites. Correlations based on variations in cal- cium carbonate content provide a third estimate of stratigraphic corre- lation between sites. These three stratigraphies enable us to estimate the precision with which sites can be correlated. In all cases, we try to account for the sampling resolution of the biostratigraphic and car- bonate data. BIOSTRATIGRAPHIC CORRELATIONS The biostratigraphic data used for correlating sites are the first and last occurrences of numerous radiolarian and diatom species. In general, we have not used da- tums based on evolutionary transitions or changes in rel- ative-abundance estimates. Such datums are more likely to be related to changes in local oceanographic condi- tions and may not represent synchronous events. Evolu- tionary transitions may also reflect differing taxonomic concepts of different investigators. The datums used in this summary, and the depth ranges in which they occur at each site, are listed in Tables 1 and 2. Each of the 91 datums was assigned a number, with the convention that a number with a zero to the right of the decimal place represents a last occurrence and a number with a 1 to the right of the decimal place represents a first occur- rence. The depth ranges listed in Tables 1 and 2 were de- termined from the two samples used to define the event. For a last occurrence, the topmost sample in which the species was found defines the base of the depth range, and the next (overlying) sample (in which the species is absent) defines the top of the depth range. Detailed stratigraphic correlations based on the bio- stratigraphic data were made by using the graphic corre- lation technique developed by Shaw (1964) and recently 695