Ecological classification and analysis of wetland ecosystems, northern Lower Michigan, U.S.A.

We describe an ecological, multifactor approach to wetland classification in which ecosystem types are identified on the basis of the simultaneous integration of physiography, climate, hydrology, soil, and vegetation. Aerial photographs and field reconnaissance were used to characterize the diversity of wetlands of the 4000-ha University of Michigan Biological Station, northern Lower Michigan. Twenty-eight wetland units, including nutrient-rich swamps, ombrotrophic bogs, and many intermediate types, were identified. Eight wetland ecosystems, composing 79% of the total wetland area, were sampled extensively and classified primarily on the basis of the major glacial landforms and physiographic features of the region. Canonical variates analysis was used to evaluate the distinctness of these physiographically determined units in relation to various biotic and abiotic variables. Wetland types were poorly discriminated by canonical variates analysis of overstory composition data; better separation among types was achieved using ground-flora vegetation, hydrology, or soil characteristics. To demonstrate the utility of the multifactor approach to applications in wetland ecology, vegetation–environment relationships were examined using canonical correspondence analysis. Patterns of ground-flora community composition across all ecosystems were related to substrate characteristics, primarily organic matter composition, in addition to water chemistry and light. The results suggest that a multifactor approach, within a landscape framework, is useful in distinguishing wetlands at local scales, particularly where differences in overstory vegetation among ecosystems tend to be masked by human-caused disturbance. However, the landform-mediated differences in various wetland characteristics that we observed argue for a consideration of landscape-level physiography in classification and management even at broader scales.