4.1 TOXICITY TESTING DATA CLASSIFICATION

4.1.1 Primary and Secondary Data Classification

The freshwater toxicity testing conducted by Environment Canada, on behalf of BC Environment, included nine separate test/organism combinations. A summary of the degree to which the studies conducted by Environment Canada/BC Environment met the data requirements for primary and secondary data is presented in Table 4.1.

Note: Acceptable lab practices are based on standardized test protocols for fish, invertebrates and plants (see Section IX.3, Appendix C)

Preferred toxicity test endpoints for primary classification for partial or full life cycle tests include effects on embryonic development, hatching or germination success, survival of juvenile stages, growth, reproduction and survival of adults
Preferred toxicity endpoints for secondary classification include those listed above for primary as wells as pathological, behavioural and physiological effects (Appendix B and C)
1 - Static test data is acceptable if concentrations did not change during the test and environmental conditions for the test species were maintained, conditions that the laboratory has stated were met (Pacific Environmental Science Center, 1998/1999)

The information presented in the above table confirms that the toxicity testing conducted by Environment Canada on behalf of BC Environment met the requirements for primary data for all tests, with the exception of the Microtox® IC₅₀ and the 72 Hour IC₅₀ algal bioassay. These tests did not meet the requirements of primary data because the manganese concentrations in the test solution were only measured at the beginning of the test (this shortcoming for the Microtox® IC₅₀ relates more to the fact that the test was of such short duration, making a second concentration measurement redundant). The first note following Table 4.1 indicates that although the tests were static rather than flowthrough, laboratory personnel stated that stable manganese concentrations did not change during the tests and the data could therefore be considered primary. The data met all requirements for secondary data.

The Stubblefield (1987). study on brown trout at three water hardnesses, the acute and chronic data from the Davies and Brinkman (1994) study on exposed and unexposed rainbow and brown trout in soft water, and the acute data for brown trout in hard water (Davies and Brinkman, 1995) were also classified using the primary and secondary data classification protocol.

Note: Acceptable lab practices are based on standardized test protocols for fish, invertebrates and plants (see Section IX.3, Appendix C)

Preferred toxicity test endpoints for primary classification for partial or full life cycle tests include effects on embryonic development, hatching or germination success, survival of juvenile stages, growth, reproduction and survival of adults
Preferred toxicity endpoints for secondary classification include those listed above for primary as wells as pathological, behavioural and physiological effects (ref. table IX-5, Appendix IX, Canadian Water Quality Guidelines, CCME 1991)
1 - Stubblefield et. al. (1997)
2 - Davies and Brinkman (1994)
3 - Davies and Brinkman (1995)

n.a. - not available, the report did not indicate whether it was static or flowthrough

The Davies and Brinkman (1994, 1995) reports did not specify whether the tests were static or flowthrough. Based on the information provided in the materials and methods sections of the referenced studies, the remaining data are considered to meet the BC Protocol requirements for primary data (and consequently for secondary data).

4.1.2 Full/Interim Guideline Classification

Literature studies completed by Stubblefield et. al (1997) and Davies and Brinkman (1994, 1995) were considered to be suitable for inclusion in the data set to be used for guideline derivation. This was based on the evaluation of the data from these studies with respect to the requirements for primary and secondary data. This literature data was combined with the new toxicity test data and the BC Protocol was applied to determine the extent to which the combined data set met the requirements for full or interim guideline development. Table 2.1 of Appendix B summarizes the minimum requirements for guideline development. A summary of the full and interim guideline requirements and an evaluation of the combined data set is presented in Table 4.3

The requirements for type and number of toxicity test were met for development of a full acute criterion and an interim chronic criterion.

4.1.3 Summary of Data Sufficiency

The new toxicity test data combined with the Stubblefield (1997) and Davies and Brinkman (1994, 1995) data did not meet the requirements for full guideline derivation for either acute or chronic guideline derivation. For both acute and chronic criteria, this was due to use of static testing procedures rather than flowthrough and the absence of information from the Davies and Brinkman (1994, 1995) acute methodology specifying whether the tests were flowthrough. As noted beneath Table 3.2, the new BC toxicity data may meet the primary data requirements and the Davies and Brinkman (1994, 1995) acute studies may have been flowthrough. As this was the only acute data deficiency, there may be sufficient information for full acute criteria derivation. For chronic criteria derivation, only one rather than two chronic studies on invertebrates were available. The available invertebrate data met the requirement for one chronic study on a planktonic species. However, a chronic study on a non-planktonic species was lacking as only LC₅₀ tests were conducted on Chironomis tentans and Hyalella azteca. No additional invertebrate studies on non-planktonic species were identified in the literature. Therefore, the available data are sufficient to derive interim guidelines but fall short of the requirements for full guideline development.

4.2 BC ENVIRONMENT TOXICITY TEST RESULTS

As discussed in Section .3.1, nominal water hardness values of 25 mg/L CaCO₃, 100 mg/L CaCO₃ and 250 mg/L CaCO₃ were evaluated as part of the testing program for some of the test/organism combinations. Replicate testing was conducted for several of the bioassays to further check the agreement of the results between replicate tests. Results of the toxicity testing program conducted on B.C. species are presented in the following sections. The data have been separated into acute and chronic results under the categories of fish, invertebrates and plants. Test results are summarized and presented in Appendix D.
4.2.1 Fish

Acute test results generated for fish at each water hardness value under study are presented in Table.4.4:

Note: Experimental concentration is based on the unverified concentration calculated by the laboratory technician

Actual concentration is calculated using ICP analysed manganese concentration on Day 0 for Replicate A
Corrected concentrations is the average of the actual toxicity values using Day 0 and final test day ICP manganese concentrations
n.a. - not available

Rainbow trout LC₅₀ concentrations were the lowest at water hardness values of 25 (measured at 47.6) and 250 mg CaCO₃/L while the coho salmon LC₅₀ concentration was the lowest value at a hardness of 100 mg CaCO₃/L. The lowest LC₅₀ concentrations were observed at a nominal water hardness of 25 mg CaCO₃/L for both species.

Actual and corrected concentrations were not determined for all tests. Manganese concentrations were apparently not determined for some tests on Day 0 (actual) and for most tests on final day (corrected). It appears that the laboratory assumed that differences between actual and true concentrations did not vary sufficiently to warrant analysis. This was supported by the coho salmon data at a water hardness of 100 mg/L CaCO₃. However, more variability was noted between experimental and actual concentrations; some concentrations were in good agreement (coho salmon at hardnesses of 25 and 250) while others were not (rainbow trout at 250 hardness).
Chronic test results on fish are provided in Table 4.5

Note: Experimental concentration is based on the unverified concentration calculated by the laboratory technician

Corrected concentrations are based on the final test day ICP manganese concentrations
n.a. - not available

EC₅₀ concentrations increased with increasing hardness, but were similar for water hardnesses of 100 and 250 mg/L CaCO₃. It is noteworthy that the minimum LC₅₀ concentrations for the acute tests were lower than the chronic values presented in Table 4.5, suggesting a less sensitive life stage used in the chronic study.

At a water hardness of 25 mg/L CaCO₃, two initial replicate tests resulted in 37.5%-45.8% non-viable organisms in the control groups, well in excess of the 10% threshold. A third replicate resulted in a corrected concentration of 14.6 mg/L.

4.2.2 Invertebrates

Note: Experimental concentration is based on the unverified concentration calculated by the laboratory technician

Actual concentration calculated using ICP analysed manganese concentration on Day 0 for Replicate A
Corrected concentration is the average of the actual concentrations using Day 0 and final test day ICP manganese concentrations
IC₅₀ - statistical manganese concentration resulting in a 50% decrease in the exposure endpoint of interest (e.g. light production for Microtox)
n.a. - not available

Daphnia magna was the least tolerant species at a water hardness of 25 mg/L CaCO₃, while Hyalella azteca was the least tolerant at hardnesses of 100 and 250 mg/L CaCO₃. LC₅₀ concentrations were observed to increase with increasing water hardness.

Actual and true concentrations were determined for the majority of tests. Actual and true concentrations showed good agreements. However, some experimental concentrations varied considerably from the actual and true values (most notably Chironomis tentans).

Chronic toxicity test data on invertebrates are presented in Table 4.7.

Note: Experimental concentration is based on the unverified concentration calculated by the laboratory technician

Actual concentration is calculated using ICP analysed manganese concentration on Day 0
Corrected concentration is the average of the actual concentrations using Day 0 and final test day ICP manganese concentrations
n.a. - not available

Corrected IC₂₅ concentrations of 5.4 and 9.4 mg Mn/L were observed at water hardnesses of 100 and 250 mg/L CaCO₃, respectively. The laboratory noted that excessive control deaths occurred at a water hardness of 25 mg/L CaCO₃ and attributed this to the softness of the test water. There was good agreement between the experimental, actual and corrected concentrations determined for the chronic D. magna testing.
4.2.3 Aquatic Plants

Table 4.8 presents the results of the toxicity testing conducted on aquatic plants.

Note: Experimental concentration is based on the unverified concentration calculated by the laboratory technician

IC₅₀ - statistical manganese concentration resulting in a 50% decrease in the exposure endpoint of interest (e.g. growth for S. capricomutum
n.a. - not available

Toxicity testing on the freshwater alga Selenastrum capricomutum was limited to a 72 hour IC₅₀ growth inhibition test at a water hardness of 100 mg/L CaCO₃. A manganese IC₅₀ concentration of 8.29 mg/L was determined.

4.2.4 Summary of Test Results

The lowest recorded manganese concentrations at which toxic responses occurred for the three water hardnesses under study are summarized in Table 4.9:

For the species under study, the results indicated that salmonids were the most sensitive species for acute exposure at water hardnesses of 100 mg/L CaCO₃ and 250 mg/L CaCO₃, while Daphnia magna was most sensitive at a water hardness of 25 mg/L CaCO₃. The sensitivity of Daphnia magna may be attributable in part to water hardness as evidenced by the 21 day chronic test results on Daphnia magna at a hardness of 25 mg/L CaCO₃. Boron was tested prior to manganese and chronic test results for boron at a water hardness of 25 mg/L CaCO₃ indicated control group mortality rates of 0% after Day 2, but 70% after Day 5. The Environment Canada Pacific Environmental Science Center aquatic toxicity laboratory concluded that the control deaths were related to the low water hardness. The test was therefore terminated and chronic Daphnia testing at a water hardness of 25 mg/L CaCO₃ was discontinued for boron and for manganese. Environment Canada laboratory personnel reported that high mortality rates in Daphnia have been observed at water hardness values of less than 50 mg/L CaCO₃ and thus, the observed mortality for the chronic Daphnia magna test was not unexpected. This may have also influenced the 48 hour LC₅₀ results for Daphnia magna, the 0.8 mg Mn/L LC₅₀ value may be due in part to water hardness, with the short duration of the test masking any contributory toxic effect of water hardness. The acute Daphnia magna result for a water hardness of 25 mg/L CaCO3 will therefore not be included in the derivation of an acute guideline.

The calculated IC₂₅ manganese concentration of 5.3 mg/L for Daphnia magna is considered to be a more effective measure of toxicity than either the LOEC or the NOEC concentrations. The LOEC and NOEC values are pre-selected manganese concentrations that are based on the concentrations chosen in the experimental design and a comparison of the exposure endpoint (i.e. survival, mobility) for the study organisms versus the control group relative to a preset level of statistical significance (usually p <0.05). By definition, the actual concentration at which an observable effect would occur must fall between the NOEC and the LOEC concentrations for the preset level of statistical significance. The IC₂₅ concentration is based on the experimental data and is an estimate of the concentration at which an adverse effect would be expected in 25% of organisms. Choosing 25% as an acceptable percentage of affected organisms is largely arbitrary and may be based more on societal values than scientific principles. However, the IC₂₅ has become widely accepted as a reasonable level of protection for aquatic organisms. In the case of the D. magna chronic toxicity test, the IC₂₅ concentration of 5.3 mg/L fell midway between the NOEC (3.6 mg/L) and the LOEC (6.9 mg/L). As the actual LOEC and NOEC must fall somewhere between 3.6 mg/L and 6.9 mg/L, the IC₂₅ value represents a good estimate of the actual NOEC/LOEC concentrations.

4.2.5 Water Hardness and Aquatic Toxicity

The test results generally show a trend whereby the manganese concentrations at which toxic responses were observed increase with increasing water hardness. This trend is apparent for most organisms studied, with the exception of rainbow trout, which exhibited higher tolerance prior to the occurrence of a toxic response at a water hardness of 100 versus a water hardness of 250 for the 96 hour LC₅₀ test. Replicate 96 hour LC₅₀ tests confirmed this result. It is not clear why this pattern emerged for rainbow trout. No data were found in the literature to support the conclusion that rainbow trout may be more sensitive to manganese when water hardness is increased from 100 mg/L to 250 mg/L CaCO₃. Similarly, there was no information to indicate whether or not the particular rainbow trout used in these experiments were sensitive to higher water hardness.

The hardness relationship was apparent for Daphnia magna for the 21 day chronic test; however, no manganese concentration was determined for a water hardness value of 25 mg/L CaCO₃ due to the unacceptably high incidence of experimental control deaths. Thus, it is probable that soft water would not constitute suitable habitat for Daphnia magna irrespective of the presence of manganese.

The 5 and 15 minute Microtox IC₅₀ values were observed to increase with increasing water hardness. These increases were most notable for the 5 minute test, with values increasing from 873 mg/L for a water hardness of 25 mg/L to 10542 mg/L for a water hardness of 250 mg/L CaCO₃, an approximate twelve fold increase. The 15 minute IC₅₀ test results increased from 73.1 mg/L to 124.3 mg/L, an increase of about 1.8 times. The results indicate the presence of a hardness dependent relationship; however, the effect of hardness would appear to decrease with increased exposure time for the toxicity endpoint under consideration (light production).