North-Central Section (36th) and Southeastern Section (51st), GSA Joint Annual Meeting (April 3–5, 2002)

Paper No. 0
Presentation Time: 8:00 AM-12:00 PM

20TH CENTURY SHIFTS IN HYDROLOGY AND FLOOD HAZARD ON THE LOWER MISSOURI RIVER FROM SPECIFIC-GAGE ANALYSIS AND STAGE INDEXING


HEINE, Reuben Allan, Environmental Resources and Policy, Southern Illinois Univ, Carbondale, IL 62901-4637 and PINTER, Nicholas, Geology Dept, Southern Illinois Univ, 1259 Lincoln Dr, Carbondale, IL 62901-4324, rheine@siu.edu

The purpose of this research was to document long-term trends in flow conveyance on the Lower Missouri River (LMoR), attempt to explain what mechanisms have caused these trends, and to update flood frequencies using the new “stage indexing” technique. Five stations along the LMoR were examined using specific–gage analysis (SGA), which is a technique that holds discharge constant to observe trends in a parameter such as stage over time. The results of SGA reveal two distinct patterns on the LMoR. For the highest flows, stages systematic rose for equal discharge volumes over the period of record. Conversely, stages associated with low discharges have declined over the period of record. The cross-over between falling stages and rising stages varied between 30% of bank-full at St. Joseph, MO and 100% of bankfull at Kansas City, MO. SGA can also be used for analyzing other time-series parameters such as cross-sectional area, flow velocity, or channel width. The results show that at three stations velocity appeared to be the dominate mechanism influencing shifting stages, whereas constriction in channel cross-sectional area drove increases in flood stages at the other two stations. Rising flood stage trends imply that large floods will occur more frequently than previously estimated. The stage indexing technique updates flood frequencies by indexing the stage data to the current year, thereby statistically normalizing the data and thus removing systematic trends. Using indexing, estimated frequencies for all flood events at all five stations analyzed were increased, sometimes dramatically. For example at Boonville, the largest flood occurring in the 70-year record was the 1993 flood with a stage of 35.16 ft. Without accounting for the long-term shifts, this #1 ranked flood would have at least a recurrence interval of 71 years. However, after indexing, the 1993 flood drops to the #4 rank and the adjusted recurrence interval for a stage of 35.16 ft is dropped to a recurrence interval of just below 10 years under present-day conditions. Inhabitants and managers of these floodplains should be aware of the rising risk of flood hazard.