about uncertainties can be maintained in an analysis. Uncertainty analysis
indicates the influence of a parameter, given the actual variation it represents, on the output variable. Sensitivity, on the other hand, is the degree to
which the model outcome depends on the variability of one parameter.
Thus, uncertainty analysis complements sensitivity analysis. Identifying the
sources of uncertainty in a model helps a user know when the limits of the
model’s applicability have been reached. Such analyses are designed to shed
light on the sources of variation of model output. Managers and decision
makers must be aware of the importance of this type of information and
be advised about how to interpret and use model results given sensitivities
and uncertainties. Although ecological models have sensitivities and uncertainties, knowledge of these sources of variation can serve to enhance the
use of a model and its results.
268
Thomas P. Maxwell et al.
uncertainties, the resulting classification can be a single category or a
range of plausible categories (Figure 14.2). The range of categories
relects the uncertainty in the input data. Results for a set of species
can be viewed together (Figure 14.3), allowing a comparison based on
the threat category as well as on the uncertainty of the results.
Figure 14.3. The RAMAS Red List result for several species. In some cases,
the status is uncertain and includes more than one threat category. In other
cases, there is no uncertainty in the status, and the species is assigned to only
one category.
indicates the influence of a parameter, given the actual variation it represents, on the output variable. Sensitivity, on the other hand, is the degree to
which the model outcome depends on the variability of one parameter.
Thus, uncertainty analysis complements sensitivity analysis. Identifying the
sources of uncertainty in a model helps a user know when the limits of the
model’s applicability have been reached. Such analyses are designed to shed
light on the sources of variation of model output. Managers and decision
makers must be aware of the importance of this type of information and
be advised about how to interpret and use model results given sensitivities
and uncertainties. Although ecological models have sensitivities and uncertainties, knowledge of these sources of variation can serve to enhance the
use of a model and its results.
268
Thomas P. Maxwell et al.
uncertainties, the resulting classification can be a single category or a
range of plausible categories (Figure 14.2). The range of categories
relects the uncertainty in the input data. Results for a set of species
can be viewed together (Figure 14.3), allowing a comparison based on
the threat category as well as on the uncertainty of the results.
Figure 14.3. The RAMAS Red List result for several species. In some cases,
the status is uncertain and includes more than one threat category. In other
cases, there is no uncertainty in the status, and the species is assigned to only
one category.