EXECUTIVE SUMMARY

The Mattabi Mine (Ontario) study is one of four field evaluations carried out in 1997 under the Aquatic Effects Technology Evaluation (AETE) Program, a joint government-industry program to evaluate the cost-effectiveness of technologies for the assessment of mining-related impacts in the aquatic environment. The other three mines studied were Dome (Ontario), Myra Falls (British Columbia) and Heath Steele (New Brunswick). Results of all four studies are summarized and evaluated in a separate summary report.

Mattabi Mines of Noranda Inc. operated a copper-lead-zinc open pit and underground mining operation and concentrator complex 80 km northeast of Ignace in northwestern Ontario. The mine was closed in 1991 and is undergoing rehabilitation; however, site runoff and seepages continue to be collected, treated and discharged to control impacts from acid rock drainage (ARD) and metal leaching. Effluent is treated with lime, polished and discharged to Bell Creek, which flows into Sturgeon Lake. Sturgeon Lake and other smaller lakes (including No Name Lake) were historically affected by ARD and loadings of zinc and other metals, and sediments in these waters are enriched with metals.

The objectives of the 1997 field program were to test 13 hypotheses formulated under four guiding questions:

1. are contaminants getting into the system (and to what degree and in which compartments)?
2. are contaminants bioavailable?
3. is there a measurable (biological) response? and
4. are contaminants causing the responses?

The hypotheses are more specific questions about the ability or relative ability of different monitoring tools to answer these four general questions about mine effect. The evaluation of tools included: sediment monitoring (sediment toxicity tests); fish monitoring (tissue metallothionein and metal analyses, and population/community indicators), and; integration of tools (relationships between exposure and biological responses and use of effluent sublethal toxicity).

Of the 13 hypotheses, 12 were tested at Mattabi as outlined in Table 1.1. The remaining hypothesis not tested at Mattabi relates to effluent chronic toxicity and its linkage to benthic and fish effects (H13). This hypothesis was deleted to optimize the fieldwork plan (for project budget).

The sediment quality triad was used as an additional means of evaluating the linkages between sediment toxicity, sediment chemistry and benthic community response (H10 and H11) in a sediment quality gradient in No Name Lake and Mine Creek Bay (Sturgeon Lake). The triad provides a more holistic means of evaluating the tools.

Study Design

The study design at Mattabi was based on lake sampling for benthos, sediment chemistry and sediment toxicity using a gradient design, including five exposure areas having different sediment zinc concentrations (No Name Lake, Mine Creek Bay) and two reference areas (Tag Lake and Peterson Cove-Sturgeon Lake). Three stations were sampled for sediment quality, toxicity and benthos within each of the seven areas. Sampling in Bell Creek followed a reference-exposure (Control-Impact) design, and allowed for testing of fish-related hypotheses. The reference area for fish collection was located in generally similar habitat nearby in a separate watershed (English River). Four stations were sampled for fish in each of the two areas, providing replicates for testing of hypotheses relating to fish populations and communities (H5 and H6).

Sampling Program

The field survey at Mattabi was completed in mid-October 1997, and included:

• water sampling at each of seven sediment/benthic sampling areas (one each) and each of two fish sampling areas (four per area). Only Bell Creek and English River data were used in hypothesis testing;
• surficial sediment sampling at each of 21 sampling stations (7 areas x 3 stations) using a petite Ponar, for determination of “total” metal concentration, partial metal concentration (i.e., the Fe and Mn oxide-bound fraction) and concentrations of acid volatile sulphide (AVS) and simultaneously extracted metals (SEM);
• a surficial sediment sampling at the same 21 stations for benthic macroinvertebrate community analysis and for sediment toxicity testing (Hyalella azteca – survival and growth, Chironomus riparius – survival and growth, Tubifex tubifex – survival and growth;
• sampling of the fish community by gillnet (at least two sets per station) at each of four exposure stations and four reference stations for determination of fish community and population parameters, and to provide specimens of white sucker and northern pike for tissue analysis; and
• sampling of 83 white sucker and 85 northern pike total from the two sampling areas for analysis of metals and metallothionein (MT) in muscle (metals only), liver, kidney and gill.

Data Overview

Water Quality

Mattabi effluent was enriched in zinc, calcium and sulphate. Bell Creek water contained elevated concentrations of zinc, copper, cadmium and lead relative to the English River, with zinc concentrations greater than Canadian Water Quality Guidelines downstream of the mine.

Concentrations of zinc were also elevated in water in No Name Lake and to a much lesser degree in Mine Creek Bay relative to the lake reference areas. Water quality guidelines were exceeded only in No Name Lake, with zinc concentration nearly two orders of magnitude above the guideline.

Dissolved and total metal concentrations showed similar spatial patterns for key metals (Zn, Cu, Cd, Pb), with dissolved and total values similar in magnitude in most cases except for Pb in No Name Lake (dissolved Pb < total Pb).

Sediment Quality

All sediments were organic-rich and sandy, with total organic carbon contents of 20 to 30%. Concentrations of total Zn, Pb, Cd, Cu, As, Ni and Hg exceeded Canadian Interim Sediment Quality Assessment Values at most exposure stations, with maximum values for zinc as high as two orders of magnitude above Probable Effect Levels (PELs). Metal concentrations followed a declining concentration gradient across the range of exposure areas sampled.

Partial metal concentration gradients were weaker than those for total metals, implying that the largest fraction of metals present is relatively strongly bound to sediment particles.

The SEM/AVS ratios showed no distinct spatial patterns, although SEM and AVS values were much greater in the exposure gradient than at reference areas. The results imply some potential for sediment toxicity, although ratios were relatively low (<1.5) in most exposure areas (except No Name Lake where higher ratios occurred). A ratio above one implies some potential for sediment toxicity, especially when SEM values are high.

Sediment Toxicity

Sediment toxicity (lethal and sublethal effects) showed little or no response in any test species to the sediment quality gradient at Mattabi. Although mortality occurred in both Chironomus and Hyalella, responses appeared similar regardless of location (reference vs exposure).

Benthic Macroinvertebrates

Benthic community trends were apparent in the sediment quality gradient, with reduced numbers of taxa in the most effected sediments. Hydracarina (water mites), Caenis (mayfly) and Pisidium (pea clam) in particular were reduced in abundance at the most affected locations.

Fish

Catches-per-unit-effort for all fish species combined were not distinctly different between reference and exposure areas. White sucker appeared more abundant in the exposure area and northern pike more abundant in the reference area. In white sucker, growth, liver size, gonad size and fecundity were comparable in reference and exposure fish. In pike, fish size at age, organ sizes and fecundity were distinctly greater in exposed (Bell Creek) fish relative to the reference area. When adjusted for body weight, organ sizes remained greater in exposed pike.

Tissue metal concentrations showed some reference-exposure differences, with differences most pronounced for selenium in all tissues and both species. This could be mine-related as selenium was detected in Mattabi effluent, but not in any other water sample.

The only metallothionein response measured was in gill and kidney of northern pike. No metallothionein responses occurred in sucker.

Hypothesis Testing

Hypothesis testing results are summarized in Table 5.2. Results of testing indicate that some of the contaminants (metals) are bioavailable, that some biological responses occurred and that contaminants appear to cause some of these responses.

Technology Evaluation

Some of the tools evaluated demonstrated a mine effect at Mattabi, whereas others did not (Table 6.2). Monitoring tools demonstrating effects or partially demonstrating effects included most water and sediment chemistry tools (except SEM/AVS), most benthic macroinvertebrate tools, and some of the fish tissue and metallothionein tools (depending on metal, tissue type and fish species). Tools that did not show responses consistent with impact included several in the fish health, fish population/community and sediment toxicity tool boxes. The limited effectiveness of some of these tools may be due to low metal bioavailability, possibly combined with the effects of natural variation. Of the monitoring tools in the same “tool box” demonstrating effects (e.g., total vs partial metals in sediments; metals vs metallothionein in tissues, etc.), major differences in tool effectiveness were not apparent at Mattabi (Table 6.3).

Overall, the relatively subdued impacts of metals at Mattabi are unexpected and noteworthy. This general condition contrasts with the greater bioavailability of metals and impact at Heath Steele and Myra Falls. This is particularly unusual because metal concentrations in Mattabi sediments are greater than observed at any of the other mines studied in the AETE program.

Conclusions on the cost-effectiveness of the tools based on results from all four mine sites studied in 1997 are found in a separate document “Summary and Cost-Effectiveness Evaluation of Aquatic Effects Monitoring Technologies Applied in the 1997 AETE Field Evaluation Program”.