Assessment factors ecosystem data with

Model ecosystem data with small assessment factor... 

It is desirable to assess whether a proposed aquatic EQS is accurate in terms of the known impacts of the chemical in the real world. This can be done by comparing the EQS with suitable model ecosystem data (if these were not used in its derivation) or better still by inspection of appropriate field data, particularly from pollution gradients for which the presence of any confounding factors (e.g., other chemicals) is well understood. 

There is no systematic bias in chemical sensitivity between freshwater and marine species, but use of freshwater data to support the derivation of marine EQSs should be conducted with caution on a case-by-case basis. Marine EQSs based on read-across from freshwater data should be regarded as tentative rather than definitive. Furthermore, the somewhat higher biodiversity in the marine ecosystem as a whole should not automatically result in the use of higher assessment factors when deriving marine EQSs. 

The Predicted No Effect Concentration may be derived from laboratory, field or theoretical data. Studies conducted on single species such as acute toxicity to fish (IX so) over a relatively short time scale (normally 40 or % h) and with death as the only recorded endpoint is, by itself, only of limited value in deciding whether or not a predicted environmental level of a dye is, or, is not, acceptable. Extrapolation from acute effects to chronic and ecosystems effects involves numerous uncertainties. In order to protect the ecosystem, conservative assessment factors have been introduced based on the statistical analysis of a set of data [17] for chronic exposure. The US-EPA [18] has proposed to apply a factor of KXX) for a single acute L(E)Cso value or 100 to the lowest value if all 3 tests are available (fish, daphniae, algae). These models have in common that they assume steady state concentrations in the aquatic environment. 

---