Using existing models and developing new mathematical models not only
elucidate important patterns and processes in natural systems, but may also
highlight the variety of challenges in addressing real-world problems facing
natural resource managers. An important challenge to effective environmental decision making is the ability to evaluate levels of uncertainty in
model predictions. Taylor et al. (2000) underscore the need to incorporate
uncertainty in management models, and this important component in ecological modeling is often overlooked.
In environmental management applications, models that do not effectively quantify or communicate uncertainty in their outputs may lead to
decisions that produce unintended results. On the other hand, these models
may simply lead to management inaction because relative risks cannot be
evaluated and, therefore, no single scenario can be demonstrated to be
better than another (Akçakaya and Raphael 1998). Hence, models that do
not adequately quantify uncertainty in their output undermine their utility
as effective management tools.
The software RAMAS Red List: Threatened Species Classifications
Under Uncertainty is a classroom example that highlights the importance
of uncertainty propagation within an applied model (Akçakaya et al. 2001).
Users can explicitly incorporate uncertainties in the input data, allowing
those uncertainties to propagate through the model, affecting the final classification of particular threatened species (see Sidebar 14.1). Such models
clearly demonstrate how the predictive power of models hinges on the
robustness of the input parameters, how managers must evaluate model
outcomes in light of such uncertainties, and how models may direct future
sampling protocols and monitoring programs.
14.2.3 Available Models and Methodologies
Modeling is a powerful tool for scientific exploration with extremely diverse
applications that span ecological models ranging from conceptual to mathematical to simulation. For example, ecological models are often simple
conceptualizations consisting of narrative descriptions, schematic diagrams, and/or box-and-arrow flowcharts. They can also consist of simple or
complex numerical equations, graphs, or computer algorithms forming the
basis of a dynamic simulation model. This latter group of models varies
greatly. Computer simulations include, for example, models of population viability, forest succession and disturbance, fires, animal habitats
and dispersion, wetland and river dynamics, and whole-ecosystem
biogeochemistry.
To increase the effective use of models in management, managers and
decision makers need to be introduced to the plethora of available models
and methods like those discussed in this book. Often, they also need to be
assisted in choosing the type of model that would best help them understand their system and guide their management decisions. In outlining and
14. Educational Investments
269
elucidate important patterns and processes in natural systems, but may also
highlight the variety of challenges in addressing real-world problems facing
natural resource managers. An important challenge to effective environmental decision making is the ability to evaluate levels of uncertainty in
model predictions. Taylor et al. (2000) underscore the need to incorporate
uncertainty in management models, and this important component in ecological modeling is often overlooked.
In environmental management applications, models that do not effectively quantify or communicate uncertainty in their outputs may lead to
decisions that produce unintended results. On the other hand, these models
may simply lead to management inaction because relative risks cannot be
evaluated and, therefore, no single scenario can be demonstrated to be
better than another (Akçakaya and Raphael 1998). Hence, models that do
not adequately quantify uncertainty in their output undermine their utility
as effective management tools.
The software RAMAS Red List: Threatened Species Classifications
Under Uncertainty is a classroom example that highlights the importance
of uncertainty propagation within an applied model (Akçakaya et al. 2001).
Users can explicitly incorporate uncertainties in the input data, allowing
those uncertainties to propagate through the model, affecting the final classification of particular threatened species (see Sidebar 14.1). Such models
clearly demonstrate how the predictive power of models hinges on the
robustness of the input parameters, how managers must evaluate model
outcomes in light of such uncertainties, and how models may direct future
sampling protocols and monitoring programs.
14.2.3 Available Models and Methodologies
Modeling is a powerful tool for scientific exploration with extremely diverse
applications that span ecological models ranging from conceptual to mathematical to simulation. For example, ecological models are often simple
conceptualizations consisting of narrative descriptions, schematic diagrams, and/or box-and-arrow flowcharts. They can also consist of simple or
complex numerical equations, graphs, or computer algorithms forming the
basis of a dynamic simulation model. This latter group of models varies
greatly. Computer simulations include, for example, models of population viability, forest succession and disturbance, fires, animal habitats
and dispersion, wetland and river dynamics, and whole-ecosystem
biogeochemistry.
To increase the effective use of models in management, managers and
decision makers need to be introduced to the plethora of available models
and methods like those discussed in this book. Often, they also need to be
assisted in choosing the type of model that would best help them understand their system and guide their management decisions. In outlining and
14. Educational Investments
269