9413134 Sparks Low frequency but high magnitude disturbance events such as severe fires or floods are key factors structuring many ecosystems (Grime, 1979; Huston, 1979; Picket & White, 1985; sparks et al., 1990). the 1993 flood on the Upper Mississippi River provides a rate opportunity to investigate the impact of such an event on the components and processes of a large river floodplain ecosystem. The recurrence interval of a flood equivalent to that of 1993, which was of exceptionally long duration (March to October: Figure1), has been variously estimated at 100-500 years by the Army Corps of Engineers and the US Geological Survey. Although it is acknowledged that nutrient supply is usually the main limiting factor governing primary productivity, the nutrient dynamics of riverine wetlands has been relatively little studied. Large floods of long duration will affect nutrient dynamics of the floodplain in a number if ways. they will deposit sediment rich in organic matter, but will also increase the rates of processes such as denitrification which will lead to a loss of nutrient from the system. Flooding will also change the vegetational composition to that of an early successional state: the rates of soil processes are strongly influenced by the species composition of the plants growing in them. There are few data on nutrient dynamics of large rive floodplains, and even less for large rivers with a substantial part of their floodplain still intact,. %%% We propose to investigate the effects of the 1993 flood in the nutrient dynamics of the floodplain by comparing the nutrient content and rates of soil processes of a range of sites which were inundated by the flood to a different extent. The 1993 flood provides a unique research opportunity. We propose to cooperate with Dr. Jos Verhoeven (Utrecht University, The Netherlands) whose group is one of the leaders in research on wetland nutrient dynamics.
