Thus, a major responsibility of the modeling task is to help decision
makers and stakeholders identify which endpoints and associated measures
can be estimated and with what uncertainty and to provide the best estimates possible (including uncertainty) of potential changes in these measures, given time and other resource constraints. Public policy decisions often
must make do with satisfactory models, not necessarily those that are scientifically optimal. Hence, decision-making constraints and context, such as
the resources available and the alternative CWIS choices being considered,
should guide model development and use.
3.3 Difficult Questions That Must Be Confronted
Each time the need arises for an assessment of the impacts of entrainment
and impingement, the same question surfaces: What modeling approach (or
specific model) should be used? We recommend that this question not be
answered until the following more difficult questions have been confronted
and addressed by all the parties involved.
1. What is at stake in terms of alternative CWIS decisions? Answering
this question involves evaluating the values and objectives of all parties.
Estimating relative decline in single populations, broader ecological
changes, or recreational or commercial-fisheries impacts may each require
different modeling approaches. Values matter because, if an outcome is
highly valued, additional or more accurate information may become more
relevant, regardless of the magnitude of physical or biological change.
Further, decision making implies comparing solutions rather than problems,
so it is appropriate to ask, “By how much do alternative decisions change
the impact(s)?” Although some minimal baseline knowledge is always
needed, modeling for decision making requires only the information
needed to compare decision options, not to produce the most realistic
model. Finally, some model outcomes may require considerable judgment
for interpretation, such as potential shifts in the trophic structure of a fish
community.
The modeler has to provide the necessary information to distinguish
between decision options, including the selection of appropriate endpoints
and associated measures and decision criteria that reflect the values of the
various parties (Keeney 1992). While, in principle, endpoints and associated
measures and criteria are distinct from value judgments, in practice some
value judgments are implicit in their selection. We recommend the use
of EPA’s framework for ecological risk assessment (USEPA 1998) as discussed below, although this framework still implies a rigid and unrealistic
3. Modeling Fish Entrainment and Impingement Impacts
53
makers and stakeholders identify which endpoints and associated measures
can be estimated and with what uncertainty and to provide the best estimates possible (including uncertainty) of potential changes in these measures, given time and other resource constraints. Public policy decisions often
must make do with satisfactory models, not necessarily those that are scientifically optimal. Hence, decision-making constraints and context, such as
the resources available and the alternative CWIS choices being considered,
should guide model development and use.
3.3 Difficult Questions That Must Be Confronted
Each time the need arises for an assessment of the impacts of entrainment
and impingement, the same question surfaces: What modeling approach (or
specific model) should be used? We recommend that this question not be
answered until the following more difficult questions have been confronted
and addressed by all the parties involved.
1. What is at stake in terms of alternative CWIS decisions? Answering
this question involves evaluating the values and objectives of all parties.
Estimating relative decline in single populations, broader ecological
changes, or recreational or commercial-fisheries impacts may each require
different modeling approaches. Values matter because, if an outcome is
highly valued, additional or more accurate information may become more
relevant, regardless of the magnitude of physical or biological change.
Further, decision making implies comparing solutions rather than problems,
so it is appropriate to ask, “By how much do alternative decisions change
the impact(s)?” Although some minimal baseline knowledge is always
needed, modeling for decision making requires only the information
needed to compare decision options, not to produce the most realistic
model. Finally, some model outcomes may require considerable judgment
for interpretation, such as potential shifts in the trophic structure of a fish
community.
The modeler has to provide the necessary information to distinguish
between decision options, including the selection of appropriate endpoints
and associated measures and decision criteria that reflect the values of the
various parties (Keeney 1992). While, in principle, endpoints and associated
measures and criteria are distinct from value judgments, in practice some
value judgments are implicit in their selection. We recommend the use
of EPA’s framework for ecological risk assessment (USEPA 1998) as discussed below, although this framework still implies a rigid and unrealistic
3. Modeling Fish Entrainment and Impingement Impacts
53