Scaling with known uncertainty: A synthesis

Abstract

Scale is a fundamental concept in ecology and all sciences (Levin 1992, Wu and Loucks 1995, Barenblatt 1996), which has received increasing attention in recent years. The previous chapters have demonstrated an immerse diversity of scaling issues present in different areas of ecology, covering species distribution, population dynamics, ecosystem processes, and environmental assessment. Scale issues occur in every facet of ecological research, including study design, data collection, experimentation, statistical analysis, and modeling. The scales of observations and outcomes in the case studies range from plots, ecosystems, landscapes, to regions. Readers will surely ask then, what new synthesis can be achieved from these and other recent contributions to the literature on scale? We see several overarching themes evident in the theory, methods, and case studies presented here, not necessarily in every chapter, but from the body of work as a whole. The following themes are illustrative: novel ideas for integrating diverse scaling perspectives, distinctions among sources of uncertainty, advances in the quantification of scaling error, improved applications of scaling principles, improved recognition of the phenomenon of scale effects (especially for cross-scale material exchange of chemicals, gases, etc.), and advances in the use of scale-related understandings for public policy and decision-making. Taken together these themes can be understood and organized by thinking through three closely related scale issues: identifying characteristic scales, understanding scale effects, and developing methods for scaling and quantifying sources of error in relation to uncertainties. In this last chapter of the book, we attempt to build from the richness of the methods and case studies toward an integration of the entire volume. To do this we briefly recapitulate scale and scaling concepts, summarize how different kinds of scale issues are dealt with in the chapters, and present a synthesis in the form of a pluralistic scaling paradigm. In the end, we conclude with some general guidelines for scaling. © 2006 Springer.