492
ized procedures (i.e., exposure media preparation, analytical chemistry reporting,
etc.). Hence, interlaboratory comparisons of test results would benefit from
agreed- upon factors including preparation method, oil loading rate, mixing energy,
dispersant- to-oil ratio, oil source, and weathering state, among others.
The large majority of aquatic toxicity data were generated via constant static or
static renewal tests. While these types of tests may not necessarily reflect the
dynamic field conditions that occur during an oil spill (i.e., water column mixing
and dilution), these types of “controlled” conditions offer one key advantage: by
allowing exposure concentrations to remain relatively constant over the test period,
body burdens buildup to levels that lead to a measureable adverse effect (i.e., death).
Thus, these types of data are useful in the development and validation of predictive
models (McGrath et al. 2018) that could then be applied with confidence in the
assessment of potential risks from oil spills. The challenge with these types of data
and associated predictive models is that their application to more realistic field conditions could only be made via assumptions (i.e., extrapolation factors). These however should be supported toxicity data, which are currently largely absent from the
scientific literature (i.e., static vs. dynamic tests).
28.5 Summary and Recommendations for Future Studies
Toxicity testing of oil and dispersants is essential in providing synthesis data to (1)
improve NRDA, hazard assessments, and operational field decisions and (2) refine
toxicity models to predict or forecast toxicity in new, understudied, or sensitive species for future oil spills. The DWH spill resulted in a massive expansion of laboratory toxicity studies in both standard and a diverse array of nonstandard species.
These results supported NRDA (DIVER 2017) and provided data to populate new
toxicity models, databases, and species sensitivity assessments (CAFE; NOAA/
ERD 2015). Many studies were of limited utility resulting from selecting inadequate methods for experimental design, chemical analyses of exposure media, and
data analysis/interpretation. Thus, it is clear that there is a need to reevaluate and
reiterate expected standards for conducting toxicity bioassays in the light of new
knowledge and experiences gained following DWH. Even despite guidelines for
conducting toxicity tests (CROSERF and modifications thereof; Singer et al. 2000;
NRC 2005), many oil, dispersant, and dispersed oil studies were not of sufficient
quality or lacked sufficient experimental details to be disregarded in a synthesis of
DWH toxicity results (Coelho et al. 2013; Redman and Parkerton 2015; Bejarano
et al. 2018). For example, in a recent synthesis of dispersant-only toxicity results,
many studies were not included as they did not meet the minimal selection criteria,
and a quarter of the studies included in that synthesis did not provide detailed information on which to assess study quality (Bejarano 2018). Table 28.2 summarizes
some of the main points to consider when conducting oil and dispersant toxicity
tests, and Fig. 28.2 highlights some of the main additional considerations post
CROSERF which were included in many DWH investigations.
C. L. Mitchelmore et al.
ized procedures (i.e., exposure media preparation, analytical chemistry reporting,
etc.). Hence, interlaboratory comparisons of test results would benefit from
agreed- upon factors including preparation method, oil loading rate, mixing energy,
dispersant- to-oil ratio, oil source, and weathering state, among others.
The large majority of aquatic toxicity data were generated via constant static or
static renewal tests. While these types of tests may not necessarily reflect the
dynamic field conditions that occur during an oil spill (i.e., water column mixing
and dilution), these types of “controlled” conditions offer one key advantage: by
allowing exposure concentrations to remain relatively constant over the test period,
body burdens buildup to levels that lead to a measureable adverse effect (i.e., death).
Thus, these types of data are useful in the development and validation of predictive
models (McGrath et al. 2018) that could then be applied with confidence in the
assessment of potential risks from oil spills. The challenge with these types of data
and associated predictive models is that their application to more realistic field conditions could only be made via assumptions (i.e., extrapolation factors). These however should be supported toxicity data, which are currently largely absent from the
scientific literature (i.e., static vs. dynamic tests).
28.5 Summary and Recommendations for Future Studies
Toxicity testing of oil and dispersants is essential in providing synthesis data to (1)
improve NRDA, hazard assessments, and operational field decisions and (2) refine
toxicity models to predict or forecast toxicity in new, understudied, or sensitive species for future oil spills. The DWH spill resulted in a massive expansion of laboratory toxicity studies in both standard and a diverse array of nonstandard species.
These results supported NRDA (DIVER 2017) and provided data to populate new
toxicity models, databases, and species sensitivity assessments (CAFE; NOAA/
ERD 2015). Many studies were of limited utility resulting from selecting inadequate methods for experimental design, chemical analyses of exposure media, and
data analysis/interpretation. Thus, it is clear that there is a need to reevaluate and
reiterate expected standards for conducting toxicity bioassays in the light of new
knowledge and experiences gained following DWH. Even despite guidelines for
conducting toxicity tests (CROSERF and modifications thereof; Singer et al. 2000;
NRC 2005), many oil, dispersant, and dispersed oil studies were not of sufficient
quality or lacked sufficient experimental details to be disregarded in a synthesis of
DWH toxicity results (Coelho et al. 2013; Redman and Parkerton 2015; Bejarano
et al. 2018). For example, in a recent synthesis of dispersant-only toxicity results,
many studies were not included as they did not meet the minimal selection criteria,
and a quarter of the studies included in that synthesis did not provide detailed information on which to assess study quality (Bejarano 2018). Table 28.2 summarizes
some of the main points to consider when conducting oil and dispersant toxicity
tests, and Fig. 28.2 highlights some of the main additional considerations post
CROSERF which were included in many DWH investigations.
C. L. Mitchelmore et al.