Relating stressor-response data from other laboratory or field studies to the case at hand often requires extrapolation. Common issues include:

• Extrapolating poorly characterized stressors.
  • Biological stressors such as invasive species and pathogens-- consider environmental requirements (habitat or hosts), critical life cycle stages and potential for acclimation
  • Physical stressors such as temperature, DO, flow - consider the difference in frequency and intensity of human caused and natural fluctuations or disturbance regimes

• Extrapolating lab to field effects. - Consider:
  • The effects of environmental conditions on chemical exposures (transformation, bioavailability)
  • The comparability of exposure circumstances (exposure routes, timing, duration)
  • The influence of avoidance behaviors and use of refugia
  • Factors influencing field populations that are not present in lab studies (exposure of critical life stages, predation, competition, other stressors)
  • Differences in sensitivity between laboratory and field organisms

• Extrapolating between geographical areas- Consider
  • Environmental conditions (climate, disturbance, geomorphology, hydrology)
  • Criticality of environmental conditions, whether they exert a major influence on the structure and function of ecological systems
  • Spatial scale and heterogeneity (the effects of habitat patch size, diversity, and distribution on connectivity and conductance)

The approaches used to address extrapolation include:

• Judgment Approaches:
  • Rely on professional expertise to relate the information that is available to observed biological impairments
  • Are essential when databases are inadequate to support empirical models and process models are unavailable or inappropriate

• Empirical Approaches:
  • Numerically extrapolate experimental or field observations to the case at hand.
  • Uncertainty factors are applied to extrapolate data from elsewhere to site conditions to ensure that the benchmark for comparison is sufficiently protective. However, protective values (e.g., criteria) are not optimal for causal analysis because it is information on the stressor intensity at which effects may occur that is required.
  • Although many data are available for chemical stressors and aquatic species, data do not exist for all taxa or effects
  • Databases for non aquatic species and most biological and physical stressors are extremely limited
  • Taxonomic extrapolations compensate for lack of data on species present at the site
  • Taxonomic relatedness is important, with extrapolations between species within genera and genera within families being more reliable than extrapolations between orders, classes, and phyla
  • Dose-scaling or allometric regression can be used to extrapolate bioaccumulation and effects of chemical stressors to another species based on relative organ masses and toxicokinetic processes (eg, renal clearance, basal metabolic rate, food consumption)

• Process-Based Approaches:
  • Representations or abstractions of a system or process that incorporate causal relationships and provide a predictive capability that does not depend on the availability of existing stressor-response information as empirical models
  • Translate data on individual effects (eg, mortality, growth, and reproduction) to potential alterations in specific populations, communities, or ecosystems
  • Using expected species abundance and population structure, single-species population models could be used to explore whether the observed population characteristics resulted from the stressors being investigated
  • Requires a thorough understanding of growth rates, physiological rates, fecundity, survival rates of the species under consideration
  • Requires knowledge of the influence of environmental factors and population density influences its biology relative to the influence of the stressor on its biology
  • Community and ecosystem models can be used to investigate direct and indirect effects on structural (eg, community composition) or functional (eg, primary production) characteristics
  • Given expected community or ecosystem properties (populations, functional types, feeding guilds, or environmental processes), estimate whether stressor could result in observed property impairment
  • Critical shifts in individual ecological components and processes can be explored using submodels that describe dynamic interactions