A NEW GENERATION OF PALEOFLOOD RECORDS FOR THE LOWER MISSISSIPPI RIVER

The lower Mississippi River is an economic artery of the United States, and extreme flooding along its course is costly in both economic and social terms. Despite ambitious engineering projects conceived in the early 20^th century to mitigate damage from extreme floods, economic losses due to flooding have increased over recent years. While these relatively moderate flood events are costly, a historically unprecedented flood capable of undermining existing flood control infrastructure would be an economic and humanitarian disaster of enormous consequence. Forecasting the risk of such an event over seasonal or longer time-scales remains a major challenge – especially in light of shifts in hydroclimatic conditions in response to continued greenhouse forcing.

Here, we present initial findings from a series of new paleoflood records that span the last 500 years derived from laminated sediments deposited in abandoned channels of the lower Mississippi River. These sedimentary archives record individual overbank floods as unique events beds with upward fining that we identify using grain-size analysis, bulk geochemistry, and radiography. Individual flood events are dated with high precision using a Bayesian age model informed by radiogenic isotopes (²¹⁰Pb, ¹³⁷Cs, and ¹⁴C) and optically-stimulated luminescence (OSL) on quartz. We use sedimentological characteristics to reconstruct flood magnitude by calibrating our records against instrumental streamflow data from nearby gauging stations.

This approach provides the first quantitative event-scale paleo-flood records for the lower Mississippi, allowing us to better understand how climate variability affects flood risk. Our paleo-flood records exhibit strong non-stationarities in flood frequency and magnitude that are associated with fluctuations in dominant modes of climate variability centered in the Pacific and Atlantic oceans. We also show that the early 20^th century was a period of anomalously high flood frequency and magnitude due to the combined effects of land use, early river engineering efforts, and natural climate variability. Extending this approach further back in time will allow us to constrain the recurrence intervals of extreme floods, and explore the influence of climate change on flood risk.