EXECUTIVE SUMMARY

The Aquatic Effects Technology Evaluation (AETE) Program commissioned a study to evaluate and summarize the experience of the Canadian mining industry with Toxicity Identification/Reduction Evaluations (TI/REs). The objectives were: i) to complete a critical evaluation of the quality of TI/RE data, its benefits and limitations and, ii) to conduct a survey to evaluate the utility of the TI/RE strategies, including discussions on TIEs and effluent treatment, in determining and/or addressing aquatic impacts from mining operations.

A review of the scientific literature indicated little pertinent published information directly related to Canadian mining and TI/RE studies and therefore the application of TI/REs to Canadian mines could not be assessed on this basis. However, the limited number of articles reviewed in this document should not give the impression that there are few evaluations being conducted, since for various reasons the overwhelming majority of studies are never published in the scientific literature.

A complete assessment of the Canadian mining sector’s experience with the TI/RE process was not possible since less than 50% of mines responded to the survey. Of 42 mines which responded, only 25 (57%) reported having experienced acute toxicity. Of those mines reporting toxic effluents, 7 (28%) indicated that a TRE had been conducted and 17 (76%) reported having conducted at least one Phase I TIE. Very few mines reported going beyond the Phase I toxicity characterization. Of the 25 mines that reported their effluents as being toxic, 9 (36%) reported that toxicity was consistent, compared with 16 (64%) that experienced transient toxicity.

Ammonia was the most commonly identified contaminant of concern and effluent toxicity appeared to be highly pH dependent. Several mines reported difficulties with the identification of secondary causes of toxicity. Five mines reported making changes to their treatment system or process based on the results of the TI/RE. Two mines reported that toxicity was eliminated following the changes, one mine reported that toxicity was reduced and two mines reported no change in toxicity. The treatment system or process changes varied from product substitution to implementation of a full-scale effluent treatment facility.

Five TI/RE case studies (CS) were also selected for detailed review. The main objective was to provide examples of TI/REs conducted by Canadian mines and the rationale behind their relative success or lack of success in implementation in terms of toxicant identification, effluent treatment changes and toxicity reduction or elimination.

In CS #1, a copper/zinc mine, the primary toxicants were identified as copper and ammonia, but secondary toxicants (silver, aluminum and total dissolved solids (TDS)) were also suspected. The TIE lead to the identification of a strategy for reduction of ammonia toxicity to rainbow trout, the main concern of the client. The mine closed, but effluent continues to be discharged and is occasionally toxic.

In CS #2, a uranium mine, the primary toxicant was identified as an aliphatic alcohol (isodecanol).  This lead to process modifications and modifications to the treatment system to solve the toxicity problem. The effluent is currently nonlethal to trout.

In CS #3, a copper/nickel mine, the primary toxicant was identified as ammonia, but secondary toxicants (metals) were suspected. The treatment system was augmented with pH control and toxicity was reduced. Mine personnel felt the results of the TIE were not worth the total cost of the studies since the conclusions were based mostly on speculation, rather than on statistically relevant results.  Treatability studies provided more relevant and applicable information, particularly related to establishing appropriate limits for pH. Historically, the effluent had been acutely lethal to trout or Daphnia magna in most toxicity tests. Most recently, the effluent has been consistently nonlethal to both species on all occasions of testing.

In CS #4, a gold mine, the general characteristics of the suspected toxicant(s) were identified (e.g., metals – most likely copper), but were not confirmed using the TIE process. Treatability investigations included bench scale evaluations, water reclamation and pilot plant studies. A full effluent treatment plant was installed based on the results of the treatability studies. However, the target level for copper used during these studies was at the Daphnia magna LC50. Effluent tested during the pilot plant trials indicated all treated samples were nonlethal to trout. However, partial daphnid mortalities (20%) were observed in those samples that exceeded the total copper target levels. During the most recent discharge period, the effluent was toxic to both rainbow trout and Daphnia magna (~ 80% mortality was observed). Ammonia, produced during the destruction of cyanide, is the suspected cause of trout mortality. Metals may be the cause of daphnid toxicity.

In CS #5, a cobalt/nickel and precious metals refinery, several possible causes of toxicity were suspected, but not conclusively identified. It was hypothesized that sodium levels were sufficient to account for at least 50% of the Daphnia magna mortality. Copper, potassium and carbonates were identified as potentially important factors in explaining daphnid mortality. Atypical ion balance was also a suspected cause of daphnid toxicity. Based on the limited available data, it was suspected that periodic peaks in sodium and/or copper concentrations contributed to the sporadic toxicity. The standard approach to toxicant identification was not possible since the U.S. EPA Phase I TIE treatments were ineffective at reducing or eliminating effluent toxicity. Subsequent toxicant identification efforts are in progress, but have required the development of innovative methodologies and techniques.

The following are the general conclusions and recommendations based on the survey responses and case studies regarding TI/REs as applied to the Canadian mining industry:

• TIEs do not “prove” the cause of toxicity, but rather use a weight of evidence approach.
• The TI/RE process is a systematic approach which incorporates the responses of organisms into the assessment of complex effluent mixtures to determine the identity of the substance(s) responsible for toxicity.
• Full transfer of information and communication between the mine and testing laboratory is critical to the success of a TI/RE study.
• The overall success of a TI/RE is based on a number of factors including the experience of the laboratory personnel performing the tests and the variability in effluent quality.
• Laboratories should be equipped with all the basic and specialized laboratory equipment required to conduct the TI/RE and the personnel should be skilled and experienced in operating the equipment.
• Beyond Phase I, the TIE approach is not standardized and subsequent studies to identify the specific toxicants require experienced personnel.
• If identification and confirmation studies are to be successful, it is crucial that the tests are well planned and scientifically defendable. It is at these stages of the TI/RE study that the experience of the investigator is crucial.
• TI/REs are generally more likely to be successful when an effluent is consistently toxic, if the loss of toxicity is minimal over time and if the factors contributing to toxicity do not vary from one sample to the next.
• Conversely, the process can be rendered more difficult if toxicity is transient, if the samples quickly lose toxicity over time or if the factors contributing to toxicity are variable (i.e. different causative agents).
• Repeated testing is required to account for effluent variability and confirm that the cause of toxicity is the same under all conditions.
• Appropriate and relevant chemical analysis should be coordinated with toxicity testing on untreated and treated effluent samples.
• The lack of statistical comparisons may not be critical at certain stages of a TIE study, where gross changes in toxicity are the only consideration. However, large amounts of data can become unmanageable and difficult to interpret without statistical analysis. Multiple regressions are likely to yield better results for matrix dependent toxicants or in cases where multiple toxicants are suspected.
• The generation of a sufficient amount of data to provide strong evidence regarding the identification of the toxicant is critical if the mine is to consider investment in costly plant-scale remedial measures.
• Toxicity testing must be included in all bench scale and pilot plant studies.
• Modifications to the standard U.S. EPA Phase I TIE approach should be investigated and specific treatment methods or approaches developed for the Canadian mining industry.
• The use of rainbow trout in a Phase I study often requires greater effort and expense since trout require large test volumes. However, surrogate test species (e.g., fathead minnows) may not be appropriate in a Canadian context. Modifications to the standard Environment Canada rainbow trout protocol should be developed and standardized for use with TIEs.