Calcareous nannofossil census data from the late Neogene of the Caribbean, the Gulf of Mexico, and the eastern equatorial Pacific were subjected to a multifold analysis in order to extract the paleoceanographic history of the region. A detailed nannofossil biostratigraphy is the basic time framework. The paleoenvironmental signals are deduced from fluctuations of diversity, dominance, and equitability through time; by clustering of assemblages through time and examining the composition of cluster groups; by comparison of fluctuations in carbonate content and coarse-fraction with time; and by factor analysis of the assemblage. These measures variously reflect surface water characteristics such as productivity, bottom water characteristics such as corrosiveness (dissolution), and conditions of sedimentation such as dilution of the pelagic biogenic carbonate. Diversity, dominance, and equitability which are determined by surface water characteristics, show similar trends, often in considerable detail, in the Caribbean and the Gulf of Mexico from the late Miocene through Pleistocene. Broad but definite similarities exist in these measures between the Caribbean/Gulf of Mexico and the eastern equatorial Pacific in the late Miocene to mid Pliocene, but not in the late Pliocene. Clustering groups assemblages in a broad biostratigraphic pattern corresponding approximately to a major assemblage zone. The structure and composition of cluster groups indicate strong similarity between the Caribbean and the Gulf of Mexico during the late Miocene through Pleistocene, but only weak similarity of the Caribbean and Gulf of Mexico to the eastern equatorial Pacific, and only during the early Pliocene. Surface water productivity is generally similar at the three locations into the late Pliocene but seemingly diverges in the eastern equatorial Pacific after about 2.4 m.y. Variations in dissolution intensity have generally similar trends in the late Miocene and early Pliocene in the Caribbean and the Gulf of Mexico but the trends diverge in the late Pliocene and are dissimilar in the Pleistocene. In the eastern equatorial Pacific, dissolution intensity increases with the inferred thermal subsidence of the crust. Variations in the calcium carbonate content of the sediment, essentially pelagic carbonate, follow different trends at the three sites, which is attributable to different sedimentary histories at the sites (tectonic plus eustatic sea level changes for the Caribbean; eustatic sea level changes alone for the Gulf of Mexico; and dilution by biogenic silica plus dissolution for the eastern equatorial Pacific).

The various measured and derived parameters indicate that as late as the mid Pliocene - approximately 3.8 m.y. - a deep water connection existed between the Pacific and the Caribbean/Gulf of Mexico. A shallower seaway, permitting exchange of surface water between the Pacific and Gulf of Mexico, seems to have existed into the late Pliocene - to between 3.0 to 2.2 m.y., and final closure probably coincided with the sea level draw-down associated with the initial buildup of a northern hemisphere ice sheet. During the late Pliocene and early Pleistocene surface productivity oscillates markedly but seemingly with different intensities in the Caribbean and the Gulf of Mexico. The most important floral change occurs in both basins at about 0.9 m.y. when the smallGephyrocapsa dominated high-productivity nannoflora of the preceeding 0.3 m.y. was replaced abruptly by a moderate-productivity flora. This turnover marks a major change in ocean circulation towards a relatively warm, stably stratified tropical surface ocean, which seems to have been affected remarkably little by some of the intense late Pleistocene high latitude climatic changes.