COMPLEXITY THEORY AND ECOSYSTEMS
Ecologist Gene Likens recently said that a major intellectual limitation for environmental studies is the false
assumption that there will be simple, all-inclusive answers (Pace and Groffman 1998). He went on to say that we
must confront the complexity of ecosystems and incorporate that complexity into our scientific endeavors.
Ecological systems are highly nonlinear, characterized by abrupt thresholds in dynamics and possibly chaotic
behavior. It is unreasonable to expect consistently accurate predictions for these systemseven with additional
resources for generating scientific information combined with the prodigious computing power now available. On
the other hand, conceptual and analytical progress is accelerating, and we can increasingly expect serviceable
forecasts of the range of likely behaviors and the probabilities of various outcomes. The key in this regard lies in
viewing systems as complex and not as the simple sum of their parts.
Ecosystem theory encompasses a wide range of approaches to understanding complex systems: Empirical work,
including experimental manipulation of natural and model systems, as well as mathematical methods drawn from
other disciplines such as cybernetics, control theory, information theory, network theory, thermodynamics, self-organization,
and emergence and hierarchy theory (Muller 1992, 1997). A fundamental issue is to integrate
systems behavior across levels of resolution in space and time to address the generation and maintenance of
biological complexity across multiple spatio-temporal levels of resolution.
Scientists have learned that even simple rules can generate very complex behaviors and that systems can be very
sensitive to initial conditions. This means that making precise long-term or large-scale predictions may be much
more difficult than we initially thoughtif not impossible in some cases. Complex systems are probably not
understandable in the same way as simple systems, although sometimes complex rules can generate simple
behavior, arguing the need to extract the "knowable" from the "unknowable" (Levin 1999). Also, small variations
may lead to large changes that are not always predictable. So-called "exceptional" events turn out to be not all
that rare. This new understanding is leading to fundamentally new approaches that will provide essential insight
and guidance to members of the public and policy-makers. Improved understanding of the behavior of complex
biological systems will greatly facilitate ecological forecasting and environmental decision-making.