possible ecological effects from those effects that are significant to the
ecosystem and/or to society (Harwell and Long 1992; USEPA 1992,
1998; Gentile and Harwell 1998). Criteria for selecting endpoints include
(1) identification of their ecological importance (e.g., important structures
and processes) or societal importance (e.g., economic or aesthetic species,
water supply, and flood protection); (2) consideration of organizational
hierarchy, including species, ecosystem, and landscape scales; (3) susceptibility to the stressors of concern; (4) identification of critical structural and
functional attributes that can be used to characterize the state (health) or
change of health of the regional environment; and (5) signal-to-noise ratio
(that is, the ability to discriminate changes in endpoints from natural
variability) (Kelly and Harwell 1990). Selection of ecological endpoints
is critical for our purposes here, because each and every selected endpoint should be addressed in an ecological risk assessment. Consequently,
the most effective and, therefore, most used, models will be those that
incorporate and provide output specific to the selected ecological
endpoints.
Problem formulation addresses the critical issue of reference or
benchmark conditions (Figure 5.3). Benchmarks are essential if scientists,
decision makers, stakeholders, and the public are to understand the
92
Mark A. Harwell and John H. Gentile
Ecological Condition
Anthropogenic
Stress Regime
Degraded
Pristine
High
Benchmarks
Lower-bound
reference
Upper-bound
reference
Current ecosystem
Desired ecosystem
Low
Figure 5.3. Benchmark, reference, and desired ecological conditions [From Harwell
et al. (1999b)].
ecosystem and/or to society (Harwell and Long 1992; USEPA 1992,
1998; Gentile and Harwell 1998). Criteria for selecting endpoints include
(1) identification of their ecological importance (e.g., important structures
and processes) or societal importance (e.g., economic or aesthetic species,
water supply, and flood protection); (2) consideration of organizational
hierarchy, including species, ecosystem, and landscape scales; (3) susceptibility to the stressors of concern; (4) identification of critical structural and
functional attributes that can be used to characterize the state (health) or
change of health of the regional environment; and (5) signal-to-noise ratio
(that is, the ability to discriminate changes in endpoints from natural
variability) (Kelly and Harwell 1990). Selection of ecological endpoints
is critical for our purposes here, because each and every selected endpoint should be addressed in an ecological risk assessment. Consequently,
the most effective and, therefore, most used, models will be those that
incorporate and provide output specific to the selected ecological
endpoints.
Problem formulation addresses the critical issue of reference or
benchmark conditions (Figure 5.3). Benchmarks are essential if scientists,
decision makers, stakeholders, and the public are to understand the
92
Mark A. Harwell and John H. Gentile
Ecological Condition
Anthropogenic
Stress Regime
Degraded
Pristine
High
Benchmarks
Lower-bound
reference
Upper-bound
reference
Current ecosystem
Desired ecosystem
Low
Figure 5.3. Benchmark, reference, and desired ecological conditions [From Harwell
et al. (1999b)].