Evidence Synthesis as the Basis for Decision Analysis: A Method of Selecting the Best Agricultural Practices for Multiple Ecosystem Services

Abstract

Agricultural management practices have impacts not only on crops and livestock, but also on soil, water, wildlife, and ecosystem services. Agricultural research provides evidence about these impacts, but it is unclear how this evidence should be used to make decisions. Two methods are widely used in decision making: evidence synthesis and decision analysis. However, a system of evidence-based decision making that integrates these two methods has not yet been established. Moreover, the standard methods of evidence synthesis have a narrow focus (e.g., the effects of one management practice), but the standard methods of decision analysis have a wide focus (e.g., the comparative effectiveness of multiple management practices). Thus, there is a mismatch between the outputs from evidence synthesis and the inputs that are needed for decision analysis. We show how evidence for a wide range of agricultural practices can be reviewed and summarized simultaneously (“subject-wide evidence synthesis”), and how this evidence can be assessed by experts and used for decision making (“multiple-criteria decision analysis”). We show how these methods could be used by The Nature Conservancy (TNC) in California to select the best management practices for multiple ecosystem services in Mediterranean-type farmland and rangeland, based on a subject-wide evidence synthesis that was published by Conservation Evidence (www.conservationevidence.com). This method of “evidence-based decision analysis” could be used at different scales, from the local scale (farmers deciding which practices to adopt) to the national or international scale (policy makers deciding which practices to support through agricultural subsidies or other payments for ecosystem services). We discuss the strengths and weaknesses of this method, and we suggest some general principles for improving evidence synthesis as the basis for multi-criteria decision analysis.