Improving the Accuracy and Precision of Predictions of TFM-niclosamide Concentrations for Treatment of Sea Lamprey Spawning Tributaries
The results of this research may lead to a more efficient use of the lampricides used to control lamprey populations in the Great Lakes. If lamprey populations are left uncontrolled, the effects on commercial and sport fisheries in the Great Lakes would be devastating.
The treatment of sea lamprey spawning tributaries with mixtures of the piscicides 3-trifluoromethyl-4-nitrophenol (TFM) and 2'5'-dichloro-4'-nitrosalicylanilide (niclosamide) is expensive and can have untoward effects. Although TFM and niclosamide are relatively selective for larval sea lamprey, juvenile lake sturgeon Acipenser fulvescens are also vulnerable (Boogaard et al. 2003). Therefore current stream-treatment protocols require application of the minimum formulation needed to kill 99.9% of the larval sea lamprey (LC99.9) while sparing juvenile lake sturgeon. Accuracy and precision are so important that stream-side bioassays have been used to estimate toxicant requirements (Howell and Allen 1962). However, the standard practice is to predict treatment requirements based on the results of laboratory toxicity tests.
The uptake of TFM decreases with increasing pH because the ionized form of the TFM molecule, which becomes increasingly dominant as ambient pH increases beyond its dissociation constant (pKa) of 6.07, transfers less readily across the gill membrane than the unionized form (Hunn and Allen 1974). This sensitivity to pH is critically important because the pH of spawning tributaries varies spatially and temporally. Additions of small amounts of niclosamide are used to boost toxicity, especially in high-pH streams. Therefore successful treatment with these agents requires prediction of toxicant formulation and concentration, based on both stream alkalinity and pH, for each application to a spawning tributary.
No single toxicity study has simultaneously controlled and varied alkalinity, niclosamide augmentation and pH over the ranges encountered by the Sea Lamprey Control Program. Bills et al. (2003) measured the toxicity of TFM alone while both varying pH and alkalinity. That study is sufficient for prediction of TFM treatment requirements, but not for prediction of concentrations of TFM and niclosamide formulations. Boogaard (unpublished data) measured the toxicity of TFM and TFM plus 1% niclosamide (by dry mass of TFM) at various combinations of alkalinity and pH. Additional niclosamide augmentation rates were not included in that design to control cost. The LC99.9 values from that study were far lower than had previously been observed. That and other evidence strongly suggests that the sea lamprey ammocoetes tested in that study were not healthy. Therefore the LC99.9 estimates from that study have not been incorporated into the current control charts. Instead, the ratios of the LC99.9 estimates from the TFM + 1% niclosamide exposures to the LC99.9 estimates from the TFM exposures have been assumed to be unaffected by health status and used to adjust the estimates of LC99.9 of TFM with respect to the 1% addition of niclosamide ( M.A. Boogaard, unpublished data). Finally, Boogaard et al. (2007) measured the toxicity of TFM and seven mixtures of TFM + niclosamide (0.125 – 2.000% dry mass of TFM) at pH values of 7, 8 and 9. That study supports prediction of the LC99.9 of broad range of TFM+niclosamide formulation from stream pH. However, alkalinity was not varied in that study, which severely limits direct application of those data to the Sea Lamprey Control Program without the additional statistical modeling proposed herein.
The current stream treatment requirements are based on informal combination of data from three toxicity studies. The nature of the informal combination precludes assessment of accuracy, precision and measures of uncertainty because it is not grounded in any statistical theory. That deficiency can be remedied. For example, Gutreuter and Boogaard (2007) demonstrated that the methods that produced the current stream treatment control charts underestimate effective concentrations, and that a more comprehensive theory of nonlinear mixed-effect models can reduce root mean-squared prediction errors by 29-82% based on a single toxicity study. However, the work of Gutreuter and Boogaard is only an insufficient first step because it did not combine data from all relevant toxicity studies and did not include the role of alkalinity in prediction. However, that work demonstrated that a statistically coherent method that combines all of the relevant historical data and a new small toxicity study should produce predictions of stream treatment requirements that are far more accurate and precise than the current stream treatment charts. The proposed combination of a new toxicity study to fill critical data gaps and Bayesian meta-analytic modeling to make full and efficient use of all of the relevant existing data should greatly improve the precision and accuracy of lampricide applications to spawning tributaries.
Objectives
- Test assumptions about the separate modes of pH and bivalent cations in toxicity of TFM and niclosamide.
- Examine the hypothesis that that large reductions in bias and large increases in the precision of predictions of stream treatment requirements are possible using Bayesian statistical models for prediction of the LC99.9 of TFM and niclosamide formulations based on the combination of the proposed and historic bioassays.
- Update stream treatment charts if there is demonstrated improvements in the precision and accuracy of LC99.9 predictions.
References
Bills, T.D., Boogaard, M.A., Johnson, D.A., Brege, D.C., Scholefield, R.J., Westman, W.R., and Stephens, B.E. 2003. Development of a Treatment Model for Applications of TFM to Streams Tributary to the Great Lakes. Journal of Great Lakes Research 29 (Supplement 1):510:520.
Boogaard, M.A., T.D. Bills, and D.A. Johnson. 2003. Acute toxicity of TFM and a TFM/niclosamide mixture to selected species of fish, including lake sturgeon (Acipenser fulvescens) and mudpuppies (Necturus maculosus) in laboratory and field exposures. Journal of Great Lakes Research 29 (Supplement 1):529-541.
Boogaard, M.A., S. Gutreuter, and C.S. Kolar. 2007. Refinement of the TFM:niclosamide pH/alkalinity prediction tables for determining minimum lethal concentrations of lampricides for applications to streams using ratios other 1.0% niclosamide in combination with TFM. Completion report submitted to the Great Lakes Fishery Commission. 25 pp.
Gutreuter, S. and M.A. Boogaard. 2007. Prediction of lethal/effective concentration/dose in the presence of multiple auxiliary covariates and components of variance. Environmental Toxicology and Chemistry 26 (9):1978-1986.
Howell, J. H. and J. L. Allen. 1962. Use of mobile bioassay equipment in the chemical control of sea lamprey. Special Science Report, Fisheries No. 418, U.S. Fish and Wildlife Service, Washingon, DC.
Hunn, J. B. and J. L. Allen. 1974. Movement of drugs across the gills of fishes. Annual Review of Pharmacology 14:47-55.
The results of this research may lead to a more efficient use of the lampricides used to control lamprey populations in the Great Lakes. If lamprey populations are left uncontrolled, the effects on commercial and sport fisheries in the Great Lakes would be devastating.
The treatment of sea lamprey spawning tributaries with mixtures of the piscicides 3-trifluoromethyl-4-nitrophenol (TFM) and 2'5'-dichloro-4'-nitrosalicylanilide (niclosamide) is expensive and can have untoward effects. Although TFM and niclosamide are relatively selective for larval sea lamprey, juvenile lake sturgeon Acipenser fulvescens are also vulnerable (Boogaard et al. 2003). Therefore current stream-treatment protocols require application of the minimum formulation needed to kill 99.9% of the larval sea lamprey (LC99.9) while sparing juvenile lake sturgeon. Accuracy and precision are so important that stream-side bioassays have been used to estimate toxicant requirements (Howell and Allen 1962). However, the standard practice is to predict treatment requirements based on the results of laboratory toxicity tests.
The uptake of TFM decreases with increasing pH because the ionized form of the TFM molecule, which becomes increasingly dominant as ambient pH increases beyond its dissociation constant (pKa) of 6.07, transfers less readily across the gill membrane than the unionized form (Hunn and Allen 1974). This sensitivity to pH is critically important because the pH of spawning tributaries varies spatially and temporally. Additions of small amounts of niclosamide are used to boost toxicity, especially in high-pH streams. Therefore successful treatment with these agents requires prediction of toxicant formulation and concentration, based on both stream alkalinity and pH, for each application to a spawning tributary.
No single toxicity study has simultaneously controlled and varied alkalinity, niclosamide augmentation and pH over the ranges encountered by the Sea Lamprey Control Program. Bills et al. (2003) measured the toxicity of TFM alone while both varying pH and alkalinity. That study is sufficient for prediction of TFM treatment requirements, but not for prediction of concentrations of TFM and niclosamide formulations. Boogaard (unpublished data) measured the toxicity of TFM and TFM plus 1% niclosamide (by dry mass of TFM) at various combinations of alkalinity and pH. Additional niclosamide augmentation rates were not included in that design to control cost. The LC99.9 values from that study were far lower than had previously been observed. That and other evidence strongly suggests that the sea lamprey ammocoetes tested in that study were not healthy. Therefore the LC99.9 estimates from that study have not been incorporated into the current control charts. Instead, the ratios of the LC99.9 estimates from the TFM + 1% niclosamide exposures to the LC99.9 estimates from the TFM exposures have been assumed to be unaffected by health status and used to adjust the estimates of LC99.9 of TFM with respect to the 1% addition of niclosamide ( M.A. Boogaard, unpublished data). Finally, Boogaard et al. (2007) measured the toxicity of TFM and seven mixtures of TFM + niclosamide (0.125 – 2.000% dry mass of TFM) at pH values of 7, 8 and 9. That study supports prediction of the LC99.9 of broad range of TFM+niclosamide formulation from stream pH. However, alkalinity was not varied in that study, which severely limits direct application of those data to the Sea Lamprey Control Program without the additional statistical modeling proposed herein.
The current stream treatment requirements are based on informal combination of data from three toxicity studies. The nature of the informal combination precludes assessment of accuracy, precision and measures of uncertainty because it is not grounded in any statistical theory. That deficiency can be remedied. For example, Gutreuter and Boogaard (2007) demonstrated that the methods that produced the current stream treatment control charts underestimate effective concentrations, and that a more comprehensive theory of nonlinear mixed-effect models can reduce root mean-squared prediction errors by 29-82% based on a single toxicity study. However, the work of Gutreuter and Boogaard is only an insufficient first step because it did not combine data from all relevant toxicity studies and did not include the role of alkalinity in prediction. However, that work demonstrated that a statistically coherent method that combines all of the relevant historical data and a new small toxicity study should produce predictions of stream treatment requirements that are far more accurate and precise than the current stream treatment charts. The proposed combination of a new toxicity study to fill critical data gaps and Bayesian meta-analytic modeling to make full and efficient use of all of the relevant existing data should greatly improve the precision and accuracy of lampricide applications to spawning tributaries.
Objectives
- Test assumptions about the separate modes of pH and bivalent cations in toxicity of TFM and niclosamide.
- Examine the hypothesis that that large reductions in bias and large increases in the precision of predictions of stream treatment requirements are possible using Bayesian statistical models for prediction of the LC99.9 of TFM and niclosamide formulations based on the combination of the proposed and historic bioassays.
- Update stream treatment charts if there is demonstrated improvements in the precision and accuracy of LC99.9 predictions.
References
Bills, T.D., Boogaard, M.A., Johnson, D.A., Brege, D.C., Scholefield, R.J., Westman, W.R., and Stephens, B.E. 2003. Development of a Treatment Model for Applications of TFM to Streams Tributary to the Great Lakes. Journal of Great Lakes Research 29 (Supplement 1):510:520.
Boogaard, M.A., T.D. Bills, and D.A. Johnson. 2003. Acute toxicity of TFM and a TFM/niclosamide mixture to selected species of fish, including lake sturgeon (Acipenser fulvescens) and mudpuppies (Necturus maculosus) in laboratory and field exposures. Journal of Great Lakes Research 29 (Supplement 1):529-541.
Boogaard, M.A., S. Gutreuter, and C.S. Kolar. 2007. Refinement of the TFM:niclosamide pH/alkalinity prediction tables for determining minimum lethal concentrations of lampricides for applications to streams using ratios other 1.0% niclosamide in combination with TFM. Completion report submitted to the Great Lakes Fishery Commission. 25 pp.
Gutreuter, S. and M.A. Boogaard. 2007. Prediction of lethal/effective concentration/dose in the presence of multiple auxiliary covariates and components of variance. Environmental Toxicology and Chemistry 26 (9):1978-1986.
Howell, J. H. and J. L. Allen. 1962. Use of mobile bioassay equipment in the chemical control of sea lamprey. Special Science Report, Fisheries No. 418, U.S. Fish and Wildlife Service, Washingon, DC.
Hunn, J. B. and J. L. Allen. 1974. Movement of drugs across the gills of fishes. Annual Review of Pharmacology 14:47-55.