Challenges and opportunities in integrating ecological knowledge across scales

Choices of the spatial and temporal dimensions of ecological investigations define their scale. In this paper, I identify some of the ways that issues of scale challenge ecologists in developing an understanding of natural and human-dominated systems, with particular reference to understanding interactions between ungulates and landscapes. I also point out opportunities to rise to those challenges. Ecologists often study areas of land that represent only a tiny fraction of the area that is managed for natural resources or other human uses. This mismatch between scales of investigation and scales of management is problematic because observations of many phenomena depend on the scale at which those observations are made. Conducting traditional experiments at ever-larger scales would appear to offer a logical solution to this problem, but the "tyranny of power" means that such investigations are frequently infeasible. Moreover, because human perception is based on limited scales of experience, it is often difficult to apply understanding of ecological processes occurring over long time periods and large areas. The ability of ecologists to integrate knowledge across scales in a way that is useful to management has improved dramatically as a result of technological advances, innovations in statistical analysis and study design, and a shift in the philosophy of science favoring model selection over traditional hypothesis testing. Multi-scale understanding is fostered by adaptive management, which uses fine-scale, mechanistic understanding to screen hypotheses to be tested at large-scales. Issues of scale reveal that applying ecological understanding to complex environmental problems requires two kinds of science-developing an understanding of properties and processes and assembling that understanding reliably across scales of time and space. (C) 2003 Elsevier Science B.V. All rights reserved.