Case study of an integrated framework for quantifying agroecosystem health

Agroecosystem health derives from a combination of biophysical and socioeconomic conditions that jointly influence such properties as productivity, sustainability, stability, and equitability. In this case study, we describe and analyze a method to quantify agroecosystem health through a combination of geographically referenced data at various spatial scales. Six key variables were hypothesized to provide a minimum set of conditions required to quantify agroecosystem health: soil health, biodiversity, topography, farm economics, land economics, and social organization. Each of these key variables was quantified by one or more attributes of a study area near Wooster, Ohio. Data sources included remote sensing, digital elevation models, soil maps, county auditor records, and a structured questionnaire of landowners in the study area. These data were combined by an analytical hierarchy process to yield an agroecosystem health index. The two steps in the process were first to combine the data at the pixel scale (30 m(2)) into key variables with normalized values, and then to combine the key variables into the final index. The analytical hierarchy process model was developed by panels of experts for each key variable and by participants in the Ohio Agricultural Research and Development Center's Agroecosystems Management Program for the final agroecosystem health index value. Observed spatial patterns of the agroecosystem health index were then analyzed with respect to the underlying data. Consistent with our hypothesis and the definition of agroecosystems, spatial patterns in the agroecosystem health index were an emergent property of combined socioeconomic and biophysical conditions not apparent in any of the underlying data or key variables. The method proposed in this study permits estimation of agroecosystem health as a function of specific underlying conditions, which combine in complex ways. Because values of the agroecosystem health index and the data underlying them can be analyzed for a particular landscape, the method proposed could be useful to policy makers, educators, service agencies, organizations, and the people who live in the area for finding opportunities to improve the health of their agroecosystem.