Agricultural land constitutes about 50 percent of the land in the U.S. and other inhabited regions. Decisions regarding the use of agricultural land have significant economic and environmental consequences. While these decisions are focused mainly upon the production of food and fiber, agricultural landscapes and watersheds also have the potential to produce enhanced flows of ecosystem services (such as nutrient cycling, regulation of atmospheric gases, soil formation, and wildlife habitat) through investments in natural capital like wetland restoration, riparian filter strips, and changes in farming methods that increase soil organic carbon. Unfortunately, there is a narrow range of socioeconomic circumstances under which resource managers are willing to make substantial private investments in the present to achieve more substantial public environmental benefits through enhanced flows of ecosystem services in the future. This interdisciplinary research project will investigate solutions to this socio-environmental dilemma by constructing a spatial decision-support system called Virtual Watershed. Virtual Watershed will combine environmental simulation models, geographic information systems, evolutionary algorithms, and agent-based modeling to capture the dynamic nonlinear interactions among economic and ecological conditions, public policies, land and water managers' behavior, and other variables that generate rural landscape patterns. These patterns in turn determine ecological and economic performance at a watershed scale. Issues of particular interest are the complex nature of landscape change over time and the capability of adaptive management processes to guide landscape change to meet user-defined social goals in an evolving environmental and social system. Case studies in two watersheds (one in Illinois and the other in Quebec) will examine whether landscape change displays self-organized complexity as captured in the concept of "panarchy," a concept associated with the idea that human-environment systems have characteristic cyclic dynamics that are linked in a hierarchy to faster and slower cycles
This project has clear and direct applications in environmental and agricultural policy. One primary hypothesis is that the environmental performance of agricultural landscapes can be markedly improved while not reducing income to land managers through changes in incentive structures. Such incentive structures include among others, forms of agricultural subsidization, land retirement programs, and cost sharing for practices adoption. Additionally, improvement can occur though the responses of land managers to other stimuli, such as neighboring land uses and neighbors' adoption of new practices. The project will further test the hypothesis that an adaptive management process that first simulates near theoretically optimal landscape performance and then is used to adjust the decision environments of land managers to guide them toward that optimum can result in improved ecological-economic performance over time. This occurs despite disturbance events and changing social preferences that have the potential to turn this adaptive management process into a treadmill of change without improvement rather than a path toward ecological-economic improvement. In testing these hypotheses, the project will achieve a better understanding of the independent and interdependent roles of economic conditions and incentives, environmental policies, resource manager behavior, and pre-existing landscape patterns on the dynamics of landscape change as well as trade-offs and complementarities among economic and ecological goals for agricultural watersheds. The visually improved Virtual Watershed model will be valuable in educational and policy-making settings. This project is supported by an award resulting from the FY 2004 special competition in Biocomplexity in the Environment focusing on the Dynamics of Coupled Natural and Human Systems.
