A paleoclimate record for the past 250,000 years from Summer Lake, Oregon,USA: II. Sedimentology, paleontology and geochemistry

We have obtained a detailed paleoenvironmental record in the Summer Lake Ba sin, Oregon (northwestern Great Basin, US) spanning from 250ka-5 ka. This r ecord is derived from core and outcrop sites extending from a proximal delt aic setting to near the modern depocenter. Lithostratigraphic, paleontologi c (ostracodes and pollen) and geochemical indicators all provide evidence f or hydroclimate and climate change over the study interval. Lithostratigraphic analysis of the Summer Lake deposits allows subdivision into a series of unconformity - or paraconformity-bound lithosomes. The unc onformity and facies histories indicate that the lake underwent several maj or lake-level excursions through the Middle and Late Pleistocene. High stan ds occurred between 200 and 165 ka, between 89 and 50 ka and between 25 and 13 ka. Uppermost Pleistocene and Holocene sediments have been removed by d eflation of the basin, with the exception of a thin veneer of late Holocene sediment. These high stands correspond closely with Marine Oxygen Isotope Stages 6, 4 and 2, within the margin of error associated with the Summer La ke age model. A major unconformity from 158 ka until 102 ka (duration varie s between sites) interrupts the record at both core and outcrop sites. Lake level fluctuations, in turn are closely linked with TOC and salinity f luctuations, such that periods of lake high stands correlate with periods o f relatively low productivity, fresher water and increased water inflow/eva poration ratios. Paleotemperature estimates based on palynology and geochem istry (Mg/Ca ratios in ostracodes) indicate an overall decrease in temperat ure from 236 ka-165 ka, with a brief interlude of warming and drying immedi ately after this (prior to the major unconformity). This temperature decrea se was superimposed on higher frequency variations in temperature that are not evident in the sediments deposited during the past 100 ka. Indicators d isagree about temperatures immediately following the unconformity (102-95 k a), but most suggest warmer temperatures between 100-89 ka, followed by a r apid and dramatic cooling event. Cooler conditions persisted throughout mos t of the remainder of the Pleistocene at Summer Lake, with the possible exc eption of brief warm intervals about 27-23 ka. Paleotemperature estimates f or the proximal deltaic site are more erratic than for more distal sites, i ndicative of short term air temperature excursions that are buffered in dee per water. Estimates of paleotemperature from Mg/Ca ratios are generally in good agree ment with evidence from upland palynology. However, there is a significant discordance between the upland pollen record and lake indicators with respe ct to paleoprecipitation for some parts of the record. Several possibilitie s may explain this discordance. We favor a direct link between lake level a nd salinity fluctuations and climate change, but we also recognize the poss ibility that some of these hydroclimate changes in the Summer Lake record m ay have resulted from episodic drainage captures of the Chewaucan River bet ween the Summer Lake and Lake Abert basins.