The amount of variation in the O2 and CO2 contents of the Phanerozoic atmosphere remain fundamental but unresolved questions. The global burial rate of organic carbon, as indicated by the carbon isotopic compositions of marine carbonates (d13Ccarb), is a critical parameter in modeling atmospheric compositions. While previous efforts have been largely focused on the redox significance of long-term, gradual trends (107 yr) in d13Ccarb and their relationship with d34S, a more volatile record that includes numerous short-term events, or isotopic excursions lasting a few million years or less, is becoming apparent. Some of the best evidence for large d13Ccarb excursions and their relationships to disturbances in the Earth's surface environment (as revealed, for example, by mass extinctions) can be found in Paleozoic strata. The proposed research is aimed at the construction of detailed d13Corg, d34Ssulfate, and 87Sr/ 86Sr for two Paleozoic time slices that are characterized by exceptionally large and well- documented d13Ccarb excursions. These Late Cambrian and Early Mississippian events have both been hypothesized to reflect global increases in the rate of organic carbon burial. This project will carry out numerical experiments using coupled box models of the geochemical cycles of C, S, and Sr to test multiple working hypotheses and to develop internally consistent scenarios for changes in the C and S cycles and the implications for atmospheric CO2 and O2 contents. Our focus on high-resolution data sets will allow us to critically examine some of the proposed cause-and effect feedback loops that stabilize the geochemical carbon cycle and thus the CO2 and O2 composition of the atmosphere. For example, burial of organic matter that is not balanced by oxidation of pyrite should increase atmospheric O2 levels (inferred from coupling and phasing of d13Ccarb, d34Ssulfate), which provides a negative feedback on organic carbon burial through removal of oceanic phosphorus associated with ferric oxides. In addition, the burial of organic matter should lower pCO2 (lowering the net difference between d13Ccarb and d13Corg), cool the climate, and result ultimately in a decrease in continental weathering rates (reducing 87Sr/ 86Sr). A more complete understanding of these two targeted Paleozoic time slices will also provide unique insight into the Earth's climate system at critical geobiological intervals during (1) a worldwide mass extinction of trilobites in the Late Cambrian and (2) the Early Mississippian spread of vascular land plants and subsequent growth of glaciers on Gondwana. Broader Impacts: The research will support graduate and undergraduate student theses at Ohio State, Missouri and Penn State, including those from underrepresented groups. Students will receive broad training in field and laboratory settings using a variety of geochemical, stratigraphic and modeling techniques. Students at Ohio State and Missouri will spend time with Kump working with box models at Penn State and join Saltzman in the field, thus facilitating exchange among student groups and exposure to diverse faculty and facilities. The researchers will also be involved in outreach activities beyond the standard dissemination of scientific information at meetings and in peer review publications. Middle-school students will be involved by coordination of fieldwork and classes with a 7th grade science teacher. The chemostratigraphic database assembled will provide key reference sections for the assessment of carbon cycling during the critical time periods. It will promote a better understanding of carbon cycling on geological timescales and its relationship with past climates and ecosystems, which will provide an important societal perspective in the context of the magnitude and impact of human-induced climate change.