Soil disturbance effects on marsh vegetation along the central Mississippi River near St. Louis, MO

Background/Question/Methods

The Riverlands Migratory Bird Sanctuary (1500 ha) was established in 1988 on US Army Corps of Engineers (USACE) land along the Mississippi river in West Alton, Missouri (38.87^o N, -90.17^o W).  It is now cooperatively operated by USACE and the National Audubon Society.  About 485 ha consist of actively managed restored bottomland prairie and marshland.  The Great Flood of 1993 caused prolonged inundation and destroyed vegetation beds and moist soil plants in the Sanctuary—an impact still visible today.  The goal of this study was to improve plant species diversity and cover, as well as generate more natural vegetation.  We employed differing degrees of soil disturbance (tilling and disking) as a means stimulate seed germination from the seed bank.  Based on related studies, we predicted that the more extreme soil disturbance, tilling, would achieve our goals best. Six replicate marshes within the Sanctuary were chosen for this study.   Each marsh was divided into three similarly sized areas and which were randomly assigned one of three treatments: disking, tilling, and control.  Vegetative cover and species presence were recorded in 0.25 m²sampling plots at random intervals along a transect in each treatment area.  A pre-treatment sample was obtained for comparison.

Results/Conclusions

ANOVA of a randomized complete block design revealed strong interactions between marsh (block) and treatment for nearly every community measure.  However, march effects for cover, species richness, and Shannon diversity were significant beyond this effect.  Treatment effects were not significant.  Community ordinations by NMDS revealed a strong tendency for plots to aggregate by marsh, not treatment.  Preliminary management recommendations can be given based on this study. Restoration activities should focus on site specific characteristics such as the species pool and local hydrology and lean away from immediate additional disturbances.  Continued observation of the vegetation response is required to validate this suggestion.