Multivariate Geospatial Analysis of Engineering and Flood Response, Mississippi River System, USA

Southern Illinois University At Carbondale, Carbondale IL

Investigators

Abstract

Pinter 0229578 From the Red River in 1997 and 2001, to North Carolina's Tar River in 1999, to the Mississippi River in 1993, 1995, 2001, and again in 2002, recurrent flooding has led to assertions that human activities have increased the magnitude and frequency of large river flooding. The hypothesis of human flood forcing, however, remains unverified and politically controversial, and a recent National Research Council panel concluded that "there are ... few cases where human impacts on flood magnitude and frequency have been carefully documented". The purpose of the research proposed here is to provide an empirical, system-wide tool for quantifying human forcing of flood stages on the Mississippi River system. We propose to document and quantify these connections using a pooled time-series analysis, a multivariate technique for testing the correlation between one or more dependent variables (here, stage magnification) and independent variables that have both cross-sectional and temporal variability. Magnification of flood hazard will be measured using specific-gage analysis of long-duration gaging stations on the study rivers. The specific-gage technique quantifies changes in the stage response to fixed discharge conditions relative to baseline historical conditions, thereby isolating the local effects of river engineering and channel modification. In order to quantify the potential independent variables, we will construct a GIS database of engineering structures and their history of emplacement and modification over the past 70-100 years on the study rivers. A database of this kind was a major recommendation of the Interagency Floodplain Management review following the 1993 flood. This database will be constructed for the analyses proposed here and then will be disseminated for widespread access and use. Local and regional engineering activities on U.S. rivers currently are being reviewed, authorized, and their relative costs and benefits assessed with little or no recognition that these activities may raise flood levels, lower recurrence times, or cause increased economic losses in future floods. The research proposed here would provide a tool for empirical assessment of the impact on flood levels of any incremental engineering addition to the Mississippi or other similar rivers. Engineers, planners, and politicians should seek to reach an equitable balance between 1) the local benefits of flood control and navigation improvements and 2) the elevated flood risk that these structures may cause.