Landscape-level classification and characterization of inland wetlands in southeast lower Michigan: Exploring HGM in a GIS

Most wetland-by-wetland classification and assessment methods currently available fail to provide resource managers with the information necessary to adequately protect and manage wetland habitats. Information concerning the wetland's surrounding landscape context, the inherent differences in wetland types across a landscape, the relative abundance of that wetland type in the area, or the ability to mitigate that wetland type successfully in order to maintain landscape diversity are often not taken into consideration. This lack of basic understanding of the characteristic differences in wetland classes and functions has led to an overall loss in wetland structural and biological diversity at the landscape-scale. To address the shortcomings of existing methods, a new landscape-level wetland classification method is needed that allows resource managers to classify and characterize all inland wetland types and potential functions across large areas in an inexpensive and efficient manner. This dissertation is designed to meet this need. Specifically, the goals of this dissertation are to: (1) explore the use of a landscape-level approach to understand and characterize site-level hydrology, chemistry, and vegetation for inland wetlands in southeast Lower Michigan, and (2) to develop a geographic information system (GIS)-based wetland classification and assessment tool resource managers can use to better manage and protect inland wetland habitats.; Components of the Hydrogeomorphic (HGM) wetland assessment method were used to classify and compare characteristics of inland wetlands in southeastern Lower Michigan. Two methods for determining HGM subclasses based on (1) the permeability of underlying geologic deposits and (2) regional ecosystems were also tested. Results showed that HGM class and subclass could be used to characterize wetland hydrodynamics (i.e., water level fluctuation), water chemistry, and related to vegetation-based wetland classes. Attempts to automate HGM wetland classification in a GIS using existing wetland polygons, however, were unsuccessful. Steps to perform this process on individual grid cells and thus meet the need of resource managers for a quick and efficient means to classify and characterize wetland types and potential functions are outlined.