Enhanced stratification and seasonality in the Subarctic Pacific upon Northern Hemisphere Glaciation-New evidence from diatom-bound nitrogen isotopes, alkenones and archaeal tetraethers

Coincident with the intensification of Northern Hemisphere Glaciation (WIG) around 2.73 million years (Ma) ago, sediment cores from both the open subarctic North Pacific and the Antarctic indicate a rapid decline in diatom opal accumulation flux to the seabed, representing one of the most abrupt and dramatic changes in the marine sediment record associated with the development of Pleistocene glacial cycles. In the North Pacific, bulk sediment nitrogen isotope data and alkenone-derived sea surface temperature (SST) estimates suggest that the productivity decline was driven by reduced exchange between surface and deep water, due to weaker wind-driven upwelling and/or a strengthening of the halocline (i.e. "stratification"). In this study of the 2.73 Ma transition at Ocean Drilling Program (ODP) Site 882 in the western subarctic North Pacific, diatom-bound nitrogen isotopes (delta N-15(db)), alkenone mass accumulation rate, and alkenone- and archaeal tetraether-based SST reconstructions support the stratification hypothesis, indicating perennially lower export production, generally higher nitrate consumption, and greater inter-seasonal variation in SST after the 2.73 Ma transition. In addition, the delta N-15(db) of large and small size fractions of Coscinodiscus spp. suggest that these diatoms grew mostly during the spring bloom during the late Pliocene, switching to their current fall-to-winter growth period at the 2.73 Ma transition; this view is consistent with their decline in dominance and provides further evidence for increased stratification (reduced vertical exchange) in the North Pacific after 2.73 Ma. The delta N-15(db) data indicate that, over the similar to 100 kyr period after the 2.73 Ma transition studied here, nitrate consumption did not reach late Pleistocene ice age levels and that nitrate consumption in post-2.73 Ma warm stages was similar to that before the transition, even though productivity was greatly reduced. We tentatively attribute this to relatively weak dust-borne iron inputs in the early post-2.73 Ma period.