SCALE ISSUES IN AGROECOLOGICAL RESEARCH CHAINS

Translating information about soil characteristics and qualities acros s different spatial and temporal scales has emerged as a major theme i n soil science. The interest in scale has developed as our understandi ng of processes operating at scales larger (e.g. landscape, regional) or smaller (molecular, aggregate) than the field plot has increased. A s next steps are considered in this area, an examination of the ecolog ical literature presents some valuable philosophical and practical con cepts pertaining to the translation of information across scales. The concepts of hierarchy and the holon are particularly relevant to the s tudy of soil as a component of an ecosystem. The experience of the eco logists over the last 30 years suggests both opportunities and constra ints for the study of soil systems at different spatial and temporal s cales. As an example, our ability to predict soil processes at the sma ll scale given large scale information (''down'' scaling) is much wors e than our ability to scale ''up''. Moreover, while there have been se veral successful efforts to scale up certain types of information, we have been unable to predict large-scale phenomena given small scale in formation in several important instances. Ecological studies provide r elevant insight. Different approaches to scale translation, and the su ccesses and failures of these different approaches, have important imp lications for soil characterization and identification of land qualiti es as we address contemporary environmental problems at different scal es. It is suggested that current scale translation efforts of all type s fail for one of two main reasons; (1) either a key controlling proce ss or characteristic has been overlooked, or (2) when multiple factors interact to create unique phenomena.