Abstract:
Flood hazard is a reality in the Southeastern U.S., which is exacerbated by the high uncertainty of flood hazard as a result of short, discontinuous stream gauge records. Extreme flood events captured in the instrumental record often represent outliers, however, paleorecords reveal these extreme floods are not as uncommon as indicated by the gauge record. This thesis used radiocarbon-dated alluvial units from published literature and unpublished archaeological reports to develop the Fluvial Activity Database for the Southeastern United States (FADSU). The database aggregates individual fluvial paleorecords along southeastern rivers in order to understand spatial and temporal patterns of flood frequency throughout the Holocene. A meta-analysis of radiocarbon samples from alluvial units was used to create chronologies of floodplain depositional activity (e.g. flooding) and floodplain stability (e.g. soil). Fluvial chronologies were built from summed probability curves that represent the likelihood of age for a radiocarbon sample created in calibration process. Sensitivity analysis of flood ages determined that chronologies are only suitable for basin-wide analysis. Thus, fluvial activity and stability chronologies were created for the Lower Mississippi Basin, Tennessee River Basin, and South Atlantic-Gulf Coast river basins. The Tennessee basin fluvial activity chronology did not show any clear correspondence with only pluvial mechanisms. The Lower Mississippi and South Atlantic-Gulf Coast basins show increased flood probability with reconstructed paleohurricanes from over-wash deposits in Lake Shelby, Alabama. Notably, when one basin demonstrated increased fluvial activity in response to land-falling hurricanes, the other did not show increased fluvial activity at the same time suggesting that the spatio-temporal nature of this meta-analysis can determine storm tracks of paleohurricanes. This thesis found two high-intensity storm tracks; hurricanes predominately tracked west through the Lower Mississippi 1,300 years ago and east through the South Atlantic-Gulf Coast 700 years ago. The connection of fluvial and coastal paleorecords in the Southeastern U.S. is a novel perspective, and based on these findings, warrants further multi-proxy investigations.
