modeling process on the part of decision makers, marketing of the sometimes poorly understood attributes by modelers, uncertainty in the model
projections, variability in the natural system, immaturity of ecological
theory, and factors that were not included in the model yet influence the
outcome of decisions.
One solution to addressing these frustrating discrepancies is to increase
interchanges between modelers and decision makers. Such interactions can
serve to improve communication and create more realistic expectations
of the contributions of models. Understanding the outcome of a model is
not achieved just by examining the graphical, mapped, or tabular output
but also by being aware of the strengths and limitations of the particular
modeling approach, the assumptions, and the uncertainties in the projections (Dale and Van Winkle 1998).
16.3.2 Defining the Problem
Careful attention to problem definition will enhance the use of models
because models designed to meet the needs of explicitly defined issues will
include the requisite elements. Such definition is not always straightforward; yet implementation of models for a particular problem often demonstrates the value of the models. That is true because implementing a model
requires explicit definition of the spatial and temporal scales of concern,
the disturbance or management actions to be considered, and explicit
hypotheses about potential interactions and effects.
Ideally, the problem definition phase involves discussions between managers and modelers. Managers have intimate familiarity with the problems
and factors that may influence them. Modelers have skills at examining
system interactions, defining key elements of an interaction, and identifying potential sources of uncertainty.The conceptual model that derives from
problem definition both guides the way that a detailed model is developed
and provides insight into the key interactions of the systems. Sometimes
this conceptual model is one of the most important products of the modeling process.
16.3.3 Using Models to Enhance Understanding
Modeling is a process that enhances the understanding of a system (Van
Winkle and Dale 1998). The process of modeling requires formulating
hypotheses about how components of a system are related and allows
exploration of the implications of those hypotheses. It identifies sensitivities and uncertainties in a system and forces ecologists to specify which
components can be considered as deterministic or stochastic.
The modeling process plays a valuable role in the overall iterative scientific process of hypothesis formulation (Overton 1977). It contributes to the
design of experimental and monitoring studies, the development and appli316
Virginia H. Dale
Précédent

- 315/327

Suivant