Soil microbial ecology of the Coosawhatchie River floodplain: Influences of microtopography, season and depth

The 'no net loss' of wetlands policy initiated in 1987 led to implementation of the wetland mitigation bank concept, under which unavoidable destruction of wetlands is mitagated by purchase of created or restored wetland credits. One major concern with mitigation banks is that areas managed to look like wetlands may not truly function as wetlands. A major obstacle to wetland functional restoration is incomplete mechanistic understanding of microbial processes in wetlands. The primary goal of this research was to assess soil nutrient pool dynamics and microbial processes in a naturally-functioning floodplain forest.; The study floodplain was on the Coosawhatchie River, a fourth order, nutrient rich, low-sediment river draining a small watershed on the South Carolina lower coastal plain. Microbial processes were studied at two microtopographic classes over two years at three replicate plot pairs. Samples were taken quarterly from three depths in the soil profile. Carbon, nitrogen, and phosphorus were the nutrients of primary interest. Microbial response variables included the total biomass pool, numbers of active bacteria, and enzyme activities. During the first year, sole-source carbon utilization ability (BiologTM) was used to evaluate shifts in soil microorganism metabolic capabilities. Redox status, pH, texture, bulk density, and water content were measured to evaluate edaphic features and conditions.; Nutrient pool sizes and ratios varied widely across the floodplain: major variations were a winter decline in root zone swale soil C, N, and P content, and overall lower C:N and C:P ratios in swale soil and litter than slope material. Higher N and P concentrations, larger clay fractions, more flooding, and lower bulk densities were generally associated with the higher nutrient-cycling enzyme activity potentials found in swales. Most enzyme activity was concentrated in surface organic material and root zone mineral soil. Correlation of nutrient pool data with microbial response variables indicated that in some cases enzyme activity responses could be partially predicted by local nutrient pool sizes, but that edaphic features and hydrologic patterns were generally more influential. Close examination of correlations between enzyme activities and variables led to the conclusion that 'hot spots' of microbial activity were associated with hydrologically-isolated, nutrient-rich microsites in swale root zone soil and macropores such as root channels in slope sub-root layer soils.