EXECUTIVE SUMMARY

This study forms part of the Aquatic Effects Technology Evaluation Program (AETE), which is a cooperative government-industry program to review and evaluate environmental monitoring technologies for the assessment of mining-related impacts on the aquatic environment. The intention is to apply sound scientific principles to environmental effects monitoring in a cost-effective manner.

The program will ultimately involve a number of studies to evaluate environmental effects monitoring technologies at a number of mine sites across Canada. Beak Consultants Limited (BEAK) was selected to undertake a pilot study in the Val-d’Or region of Quebec to guide the design of the programs that will be implemented at selected mine sites across Canada. The following environmental monitoring methods were selected by the AETE Technical Committee for evaluation at the Val-d’Or pilot site:

• a comparison of surficial sediment mapping techniques (sub-bottom acoustical profiling system versus sonar and grab sampling);
• a comparison of the effectiveness of coring devices (Hornbrook, Kajak-Brinkhurst, and alpine corers) to quantify pre-operational metal levels;
• an evaluation of the cost-effectiveness of using the lowest method detection limits achievable for water and sediment chemistry analyses;
• a comparison of analytical methods for assessing sediment quality to predict biological effects by measuring either total metals in sediments by full extraction or those readily extractable (assumed to be more biologically available) by partial digestion;
• an assessment of three sediment toxicity methods (Microtox™, Hyalella azteca, and Tubifex tubifex) to assess their ability to accurately predict adverse biological effects from metal contamination of sediments;
• a comparison of the ability of artificial substrates versus grab samples to measure mining-related effects on benthic invertebrate communities in depositional habitats;
• an evaluation of the cost-effectiveness and sensitivity of benthic invertebrate processing methods (sieve sizes, level of taxonomy, number of replicates) to delineate and monitor mining-related impacts;
• an assessment of the effects of mining discharges on fish communities by evaluating the four main response characteristics; age structure, growth, energy storage and reproduction;
• an evaluation of metal accumulations in various fish tissues (gill filament, kidney, liver, flesh, viscera); and
• a comparison of metal levels in tissues to metallothionein levels and histopathology.

The sensitivity, cost effectiveness and appropriateness of each method was assessed during the pilot study.

The study area was situated in the headwaters of the Kinojévis River, immediately west of the Town of Cadillac and approximately 40 km east of the Town of Rouyn and involved assessments in the receiving environments of Mine Doyon, Complexe Bousquet and Mine Dumagami. All three mines are primarily gold mining operations. The mine effluent receiving waters were rivière Noire, rivière Bousquet, lac Chassignolle and lac Preissac.

There were three main phases in the 1995 Pilot Study as follows:

• Phase I was an assessment of the adequacy of the site selected by the AETE Technical Committee by undertaking a field reconnaissance survey and preparing a field study design;
• Phase II was the implementation of the study design; and
• Phase III was the preparation of a final report assessing and comparing the methods and recommending modifications to the design that would allow the AETE Technical Committee to reliably compare methods in subsequent years.

COMPARISON OF SURFICIAL SEDIMENT MAPPING TECHNIQUES

This component of the pilot study included a comparative evaluation of sub-bottom acoustical profiling methods and more conventional bottom grab sampling combined with the use of a standard sonar unit.

Sediment characteristics were identified in more detail by sub-bottom profiling than by sonar/grab sampling methods. However, results of the sub-bottom profiling work in this study were far less cost effective for simple mapping of habitat types for an environmental effects monitoring study. Confirmatory cores with up to almost 1-meter penetration in some cases suggested that the sub-bottom profiling was prone to misinterpretation of sediment types. In terms of bathymetry mapping, both the sub-bottom profiler and conventional sonar yielded comparable results.

It was concluded that bottom characterization using conventional sonar techniques, supplemented with grab sampling, is generally the more cost-effective approach for identifying depositional areas and determining where sediment geochemistry and bioassessment studies should be carried out.

EVALUATION OF SEDIMENT CORING METHODS

Three coring devices were compared: Hornbrook, Alpine and K-B corers.

The study concluded that either the Hornbrook sampler or other core samplers could be used in environmental effects monitoring, however, the gravity type corers allowed for more detailed measurement of metal profiles, including knowing the exact depths at which sediment contamination extends. They also had less risk of contamination of the deeper sediments during collection when compared to the Hornbrook sampler.

Water Quality

Water quality conditions were evaluated in the rivière Bousquet/lac Chassignolle and in the rivière Noire system principally to define conditions that may influence biological communities (i.e., fish and benthic parameters). Information is also presented on costs associated with achieving relatively routine and lower analytical detection limits (1/10^th CCME water quality guidelines).

It was found on the rivière Bousquet receiving environment that concentrations of potentially toxic heavy metals (Pb, Zn, Cu, Cd) did not change appreciably between the upstream reference station and the river mouth. Concentrations of heavy metals were low in mine effluents, and were at most only about four times greater than found at the reference station. Trace element analyses of rivière Noire and ruisseau Dormenan showed several parameters elevated above upstream reference levels with arsenic being the most evident.

In terms of parameters and detection limits for trace elements the ICP-MS provided for analysis of more elements than ICP, and typically provided for detection limits which are 1,000 fold lower than provided by ICP for similar costs. Thus, ICP-MS solves detection limit problems with ICP, which typically requires supplementary analyses to achieve detection limits equal to or lower than some water quality criteria.

Sediment Quality

Sediment quality conditions were evaluated for the rivière Bousquet/lac Chassignolle system and more so in the rivière Noire system. Results showed no obvious increases in metal concentrations in lac Chassignolle relative to lac Bousquet (reference). Arsenic concentrations were highly elevated in rivière Noire and the river mouth area. The data indicated that the sources were from inactive mines. Other metals appearing to be slightly enriched downstream of the mine sites included nickel, copper and zinc.

Partial extraction results were compared to total extraction results for selective stations. Partial extraction generally extracted most of the cadmium, little of the copper, about half of the arsenic and zinc, and about 30% of the iron and nickel.

Principal Components Analysis demonstrated that station groupings based on full and partial extraction were similar and both were similar to station groupings based on benthic communities. The study concluded that full extraction of metals coupled with TOC and grain size analyses was better able to detect significant differences with greater power among reference, near-field and far-field areas.

SEDIMENT TOXICITY TESTING

Sediment toxicity was evaluated by exposing organisms to sediment and measuring survival and growth in the amphipod Hyalella azteca, survival and reproduction in the tubificid oligochaete Tubifex tubifex and inhibition of photoluminescence in the Microtox™ bacterium Vibrio fischeri.

Survival of amphipods at the reference sites ranged from 66% to 92% while the sediment toxicity downstream of the mine discharge ranged from 0 to 16%.

Survival of Tubifex tubifex at all stations was comparable to the reference sites with the exception of one station in lac Preissac. The number of hatched young produced by oligochaetes exposed to sediment at all stations was comparable to that at reference sites except at one station downstream of an idle mine site. At this location, the number of young counted per adult was also lower than reference site production by about 99%. Oligochaetes exposed to sediment downstream of the mine discharge exhibited an 83% reduction in young produced per adult and showed toxic effects approximately 1 km downstream (32% to 42% reduction in young production).

Sediment from all stations on rivière Noire showed some level of inhibition with the Microtox test when compared to IC50 values for reference stations.

Only Tubifex tubifex results showed significant differences between the reference area and the near-field. Percent Hyalella survival was the only test which showed a significant difference between reference and far-field areas.

Hyalella survival appeared to provide the most sensitive response. The Tubifex production of young per adult provided slightly less sensitivity than Hyalella survival but did provide the best graded response to environmental stress and showed less variability among reference stations. The Microtox™ test identified the most toxic sediment samples and sediments of intermediate toxicity but revealed no significant difference between reference, near-field and far-field areas.

BENTHIC INVERTEBRATE COMMUNITY ASSESSMENT

The study compared the effectiveness and sensitivity of the following benthic invertebrate processing methods:

• family, genus and species level identifications;
• sieve sizes, 200 µm, 500 µm, and 1000 µm ;
• testing with pooled and unpooled replicate data;
• artificial substrates versus grab samples in depositional habitats; and
• effect of the number of replicate samples.

The study also examined the relationships among sediment chemistry, toxicity and benthic invertebrate community structure.

The following are the major conclusions of this study:

• Unpooled replicate samples allow for characterization of variation of stations within groups, but had little additional value;
• Taxonomic identifications were influenced by sieve size;
• 200 µm mesh sieves distinguished area differences most reliably;
• Species or genus-level taxonomy distinguished area differences most reliably;
• Ponar-grab samples appeared more likely to detect differences between reference and exposed benthic communities than were suspended artificial substrates;
• Full extraction of sediment metals showed more reliable separation of reference and impacted areas but was more poorly related to benthic indices than was the case for partial sediment metal extraction.
• Detailed benthic invertebrate data (e.g., species level taxonomy and 200 µm mesh sieves) showed stronger relationships with sediment chemistry than coarse benthic data (e.g., family level taxonomy and 1000 µm mesh sieves).
• Agreement between benthic indices and toxicity endpoints was most notable for Hyalella mortality and Tubifex reproduction tests, with either detailed or coarse benthic data sets.

FISH SURVEY

The fish survey was carried out to evaluate various methods for measuring mining-related fish community impacts. The survey included the collection of information on the general health, condition and reproductive status of fish from various environments, and the collection of information on metal bioaccumulation in various tissues (muscle, gills, liver, kidney) and metallothionein (MT) responses in those tissues. Also, samples of various tissues were collected for histopathological analysis. The fish examined for this study included walleye, white sucker, silver redhorse, northern pike, yellow perch, stickleback, and brown bullhead.

The results of the study showed that on the rivière Bousquet system clear differences in concentrations of heavy metals did not exist between exposure and reference areas for individual fish species and tissue types, and MT concentrations showed no near clear upstream-downstream differences in the tissues sampled that might be attributed to mining-related metal exposure or to increased metal concentration downstream of the mines. Several metals varied with age of fish, depending on tissue type, although all of these relationships were weak and metals were significantly elevated in some tissues of some exposed fish, and in other cases in some tissues of some reference fish.

Muscle was rarely the tissue where the greatest level of accumulation occurred; kidney, gill or liver were usually dominant among tissue types in terms of metal concentration.

As observed with adult fish, metal data for small fish were highly variable in terms of differences within and among metals, species and locations.

Regression analysis of the metal and MT data showed significant and positive relationships between metals and MT in liver of adult white sucker and northern pike, although none was found in other tissue types.

Histological analyses of gills demonstrated no serious impairments in any of the fish examined. No serious impairments to kidney function were observed in white sucker and northern pike from the Bousquet-reference and Bousquet-exposure sites. Histological analyses of liver from white sucker and walleye from rivière Noire mouth demonstrated some apparently adaptive responses. These included larger hepatocyte nuclear diameter and greater hepatocyte cell areas. Northern pike from the rivière Bousquet exposure site had considerably less glycogen than liver from rivière Bousquet reference pike. Information was also obtained on vitellogenesis (egg development) in female white sucker, northern pike and walleye. No differences were observed between rivière Bousquet reference and exposure sites in egg development of white sucker and northern pike as measured by gonadosomatic indices (GSIs), relative fecundities (eggs/gram of fish), egg diameters and weights, and percent of clutch oocytes.

The major conclusions of the fisheries component of the study are as follows:

• metals may be readily measured at detectable levels in all tissue types sampled;
• there is probably some inherent redundancy in the analysis of four tissue types in adult fish;
• hepatic MT levels covaried with the sum of hepatic cadmium, zinc and copper;
• monitoring of MT in liver may be more effective than monitoring of MT in gill or kidney;
• histological work indicated some possible differences in fish tissues among sites, suggesting that such measurements may be of some value as measures of biological response; and
• young fish are easier to collect and maintain in viable condition than adult fish; field costs could be substantially reduced if small fish viscera could be monitored rather than organs of large fish.

The study has also made a number of recommendations for future AETE programs at selected mine sites across Canada.