SEARCH

SEARCH BY CITATION

DEAR SIR:

We gratefully acknowledge the response of Van der Vliet et al. (this issue) to our article on the reasons for the slow progress on the transition to alternative measures to the no observed effect concentration (NOEC)/lowest observed effect concentration (LOEC). As a Debate and Commentary article, we feel our contribution is already fulfilling its purpose (van Dam et al. 2012). In response to the letter by Van der Vliet et al. (this issue), we have chosen to highlight a number of points, both in agreement with, and contesting, their key views.

  • 1.
    We fully concede that our article unfortunately overlooked the efforts made by Environment Canada in addressing this statistical issue in its Biological Toxicity Testing Methods Series. Environment Canada's efforts are commended and represent the type of action that we are promoting, not by just 1 or 2 jurisdictions, but internationally. Unfortunately, we were unaware of these efforts (on this, refer to point 2, below). Importantly, we did not “ignore Canada,” having acknowledged the efforts made by CCME (2007) in terms of addressing limitations of NOEC/LOEC data for deriving water quality guidelines. However, our main argument “…that the major reason for the continued generation and publication of NOEC/LOEC data is that key government and inter-governmental bodies have been ‘sitting on the fence’ on the issue for more than a decade…” still stands, as Environment Canada is clearly in the minority when it comes to actual action since the mid–late 1990s to address this issue.
  • 2.
    Van der Vliet et al. (this issue) highlighted how they did not make a fanfare of their efforts to exclude NOEC/LOEC calculation in favor of regression-based methods for the statistical analysis of ecotoxicity data. We wish they had; first, we would have known about it, and second, greater international awareness of such efforts over the past decade (including the International Organization for Standardization of its efforts) may have resulted in there being broader action to date. On important issues, it is often necessary to make oneself heard. In our view, quietly changing the test protocols by inserting a footnote of the decision not to use NOECs and LOECs will, unfortunately, not silence the debate. Indeed, a bit of “arm waving” over the improvements to the Canadian toxicity test methods would have contributed much to the wave of positive momentum on this issue that Van der Vliet et al. (this issue) themselves clearly support.
  • 3.
    Van der Vliet et al. (this issue) claim that we confused environmental quality guidelines and laboratory toxicity testing. We agree that the statistical analysis instructions in the laboratory toxicity testing methods need to be changed. Such changes will result in better data being included in data sets used for guideline derivation. However, the toxicity testing methods and water quality guideline derivation methods are critically related to each other, and the latter can play an equally important role in educating the private, government, and research sectors on appropriate analytical methods. For example, as noted in our article, the current revision of the ANZECC/ARMCANZ (2000) Australian and New Zealand water quality guidelines will exclude all NOEC data from a data set once preferred toxicity estimates (i.e., ECx and no effect concentration [NEC]) are available for greater than or equal to 8 species from 4 taxonomic groups. For anyone wanting their data to be used to derive guidelines, it will be clear that they need to avoid focusing on NOEC/LOEC data. In contrast, as CCME (2007) contains no such impediment to the continued use of NOEC data for guideline derivation (other than stating a preferred hierarchy of toxicity estimates), new NOEC data for relevant species not assessed using the CCME toxicity testing methods will continue to be admissible to derive guidelines. Hence, the statistical guidance needs to be consistent and unequivocal in both the test methods and the guideline derivation guidance documents to effect the necessary change in data analysis practices.
  • 4.
    Van der Vliet et al. (this issue) also point to the US Environmental Protection Agency's (USEPA) recently developed Test of Significant Toxicity (TST) for overcoming the limitations of NOECs/LOECs in whole effluent toxicity testing (USEPA 2010). Indeed, this appears a useful approach for the objective of determining whether an effluent concentration or site water sample is truly toxic but is less relevant to the broader generation of toxicity data for individual chemicals and associated objectives of deriving water quality guidelines, for which we believe concentration–response modeling is a more relevant approach.
  • 5.
    On the issue of broader hypothesis testing applications, we agree with Van der Vliet et al. (this issue). Our article specifically focused on the need to replace the NOEC/LOEC, not all hypothesis testing. Hypothesis testing approaches remain useful for certain experimental designs, so long as the experiments are conducted properly and the analyses are appropriately qualified (as per Newman 2008).

Ultimately, it is clear that Van der Vliet et al. are working toward a similar outcome as we and others (e.g., Landis and Chapman 2011; Jager 2012), rather brutally referred to as arm wavers, are promoting. It seems there is more being done at present than our article may have acknowledged and that is pleasing to us, although we would obviously have preferred to have highlighted it in our article as a positive example of the way forward. However, the analysis by Jager (2012) and our observations arising from recent scientific forums that enormous amounts of NOEC/LOEC data are still being generated (that will be used subsequently for guideline derivation, including in Canada if the species are considered relevant), suggests that there is still a long way to go. Yes, we did point the finger, including the lack of progress in Australia, and, in an international context, we still feel that our argument is valid.

REFERENCES

  1. Top of page
  2. REFERENCES
  • [ANZECC/ARMCANZ] Australian and New Zealand Environment and Conservation Council/Agriculture and Resource Management Council of Australia and New Zealand. 2000. Australian and New Zealand guidelines for fresh and marine water quality. National Water Quality Management Strategy Paper No 4. Canberra, Australia: Australian and New Zealand Environment and Conservation Council and Agriculture and Resource Management Council of Australia and New Zealand.
  • [CCME] Canadian Council of Ministers of the Environment. 2007. A protocol for the derivation of water quality guidelines for the protection of aquatic life 2007. In: Canadian Environmental Quality Guidelines, Winnipeg, MB, Canada.
  • Jager T. 2012. Bad habits die hard: The NOEC's persistence reflects poorly on ecotoxicology. Environ Toxicol Chem 31: 22229.
  • Landis WG, Chapman PM. 2011. Well past time to stop using NOELs and LOELs. Integr Environ Assess Manag 7: viviii.
  • Newman MC. 2008. What exactly are you inferring? A closer look at hypothesis testing. Environ Toxicol Chem 27: 10131019.
  • [USEPA] US Environmental Protection Agency. 2010. National Pollutant Discharge Elimination System Test of Significant Toxicity Technical Document. Washington (DC): USEPA. EPA 833-R-10-004.
  • van Dam RA, Harford AJ, Warne MSJ. 2012. Time to get off the fence: The need for definitive international guidance on statistical analysis of ecotoxicity data. Integr Environ Assess Manag 8: 242245.
  • Van der Vliet L, Taylor LN, Scroggins R. 2012. NOEC: notable oversight of enlightened Canadians. A response to van Dam et al. (2012). Integr Environ Assess Manag 8: 397398.