Skip to main content
ARS Home » Northeast Area » University Park, Pennsylvania » Pasture Systems & Watershed Management Research » Research » Publications at this Location » Publication #287456

Title: Optimizing ecosystem function by manipulating pasture community composition

Author
item Goslee, Sarah
item Veith, Tameria - Tamie
item Skinner, Robert
item Comas, Louise

Submitted to: Basic and Applied Ecology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 9/23/2013
Publication Date: 12/1/2013
Citation: Goslee, S.C., Veith, T.L., Skinner, R.H., Comas, L.H. 2013. Optimizing ecosystem function by manipulating pasture community composition. Basic and Applied Ecology. 14:630-641.

Interpretive Summary: In pastures, important ecosystem functions include producing forage biomass and excluding weedy species. Each species planted in a pasture is effective at providing some ecosystem functions but poor at others. Relating plant characteristics to performance in grazed field plots makes it possible to predict ecosystem function so that pastures can be designed to meet particular management goals. Mathematical optimization methods make it possible to identify the best mixture for a given ecosystem function, and to explore the trade-offs between different functions. Data from greenhouses studies and field experiments were used to evaluate optimization methods for this purpose, and to identify areas where further work is needed before results can be used to make management recommendations.

Technical Abstract: The ability to design plant communities to provide particular ecosystem services would benefit agriculture, environmental mitigation, and ecosystem restoration. We propose a quantitative multi-step process to select mixtures of plant species to meet ecosystem objectives: collecting trait data; choosing suites of traits related to relevant physiological processes; relating those processes to the ecosystem functions of interest; and performing multi-criteria optimization. This process was tested using trait-based process scores for fourteen species found in pastures of the northeastern United States. Generalized additive modeling using grazed small plot data was used to relate community weighted mean process scores to timing and production of forage and nonforage (weedy) biomass and cover, and to the percentage of bare ground. Species mixtures best suited to maximize forage, and minimize nonforage species and bare ground were identified. Multi-criteria optimization enables trade-offs in processual ability within species and trade-offs in ecosystem services at the community level to be quantified, offering practical benefits to managers and improving understanding of functional ecology.