9725331 House The connection between floods and decadal-to centennial-scale climatic variations in the past has tremendous implications for probability assessments of the magnitude and frequency of future floods, especially in relation to scenarios of future regional or global climate change. This study will use paleoflood data to compare the influence of decadal-to millennial-scale climatic variability on the magnitude and frequency of large floods on rivers in three district hydroclimatic regions of the western U.S: the Southwest, Northwest, and western Great Basin. The paleoflood studies will provide physical evidence of individual flood events in the past, extending the record of severe floods on each river by 100s to 1000s of years through stratigraphic analysis of flood deposits. Previous studies in the Southwest and preliminary field work conducted during the summer of 1997 on the John Day River in Oregon and the Truckee River in Nevada confirm the availability of excellent paleoflood records from rivers in all three regions. Paleoflood chronologies on rivers in the southwestern U.S. indicate a connection between regional climatic conditions and centennial-scale variations in the frequency of large floods over the last 5000 years. A principal aim of this project is to determine whether demonstrated regional differences in the timing and controls on floods at the short time scales of the modern records hold true for the regional responses of floods to longer-term climatic variations. This broad, new perspective on flood hydroclimatology is achievable even with the relatively coarse resolution of radiocarbon dating methods. With the collaboration of Kelly Redmond, Deputy Director and Regional Climatologist for the Western Regional Climate Center, modern discharge, meteorological and atmospheric-pressure data will be analyzed to serve as analogs for the paleofloods in each region and thus aid in understanding the underlying physical bas es for the patterns observed in each regional flood series. The study will also take advantage of recent refinements in statistical computer programs for incorporating paleoflood information into single-site flood frequency analyses (developed by the Bureau of Reclamation), and methods for the quantification and reduction of paleoflood discharge-modeling uncertainties (House, 1996). Understanding links between climate and floods from the basis of real information on the magnitude and frequency of great, rare floods will improve the ability of water-supply managers to foresee the regional consequences of glodal-and regional-scale climatic phenomena, an increasingly important component of long-term planning in the face of uncertain future climate conditions.
