Warning This Web page has been archived on the Web.

Archived Content

Information identified as archived on the Web is for reference, research or recordkeeping purposes. It has not been altered or updated after the date of archiving. Web pages that are archived on the Web are not subject to the Government of Canada Web Standards. As per the Communications Policy of the Government of Canada, you can request alternate formats on the Contact Us page.

Skip booklet index and go to page content

A Decade of Research on the Environmental Impacts of Pulp and Paper Mill Effluents in Canada (1992-2002)

3.7 Evolution of the Research Questions : Cycle 2 EEM Results, What Were the Major Response Patterns and How Widespread Were They?

Industry and the research community now agreed that reproductive effects had been demonstrated downstream of some mills at numerous places around the world and that partial recovery had occurred at a number of locations following process and treatment upgrades. How widespread the responses were however was still at question. Results from Cycle 2 of the Canadian EEM program were received by Environment Canada in April of 2000. This represented the first collections following the installation of secondary treatment at the majority of the mills in Canada and would provide a first glance at the responses of the receiving environments across the country. Cycle 1 data were used as a baseline to gain a better understanding of the variability of the field measurements. This lead to an improvement of Cycle 2 study designs and gave some insight into the typical response patterns of fish and benthos to pulp and paper mill effluents. Cycle 2 proved to be the first EEM cycle with consistently good data, allowing Environment Canada to begin addressing the fundamental objective of the EEM program: to determine the potential effects of pulp and paper mill effluents on fish populations and benthic invertebrate communities. Environment Canada identified a number of questions for data analysis based on priorities set forth by the department, the pulp and paper industry and other partners.

A national assessment of EEM data was conducted with the objective of presenting the current state of aquatic ecosystems around Canadian pulp and paper mills and to present any discernible trends in the receiving environment over the past 10 years. In doing so, the report presents in depth information on the state of the fish populations and the benthic invertebrate communities existing in the receiving environments across Canada. More specifically, the report addresses a number of priority areas identifed by government EEM scientists through consultation with industry such as how the magnitude of effects and sublethal endpoints are influenced by mill process, effluent treatment, and habitat/community type of the receiver.

This assessment examined the national response patterns demonstrated in both the benthic invertebrate communities and the fish surveys. The magnitude and direction of effluent-associated effects on benthic invertebrate communities varied significantly among different habitat types. Abundance was significantly increased in both erosional and depositional riverine habitats. Taxon richness was not significantly changed when considering riverine habitats as a whole (Lowell et al., 2003). Further subdividing the riverine data set revealed that the distribution of individual mill responses for taxon richness was weighted toward mills showing significant increases and those showing no significant change. Thus, the average response pattern observed in riverine exposure areas was one of mild to moderate eutrophication. In contrast, both abundance and taxon richness decreased in marine/estuarine habitats, a response that was indicative of toxicity or a smothering response pattern (smothering effects are the result of the deposition of pulp fiber mats). The distribution for lacustrine habitats was intermediate between that for riverine and marine/estuarine habitats (Lowell et al., 2003). Mild to moderate eutrophication effects have been well documented in smaller-scale studies for mills discharging into riverine habitats (Hall et al., 1991; Culp et al., 2000; Lowell et al., 2000), and riverine habitats comprise the majority (two-thirds) of the mills included in the invertebrate analyses for Cycle 2. One of the advantages to the large-scale meta-analytical approach reported was that it had identified differences between freshwater and marine responses that were not previously documented. This led to a variety of new testable hypotheses on habitat-related effects: historical deposition of fiber mats (are fiber mats flushed less regularly in marine than in freshwater habitats), species sensitivity (are there differences among habitat types in taxa sensitivities to effluent exposure), multiple co-occurring effects (are freshwater inhibitory effects more often masked by eutrophication), and study design (is there a difference in positioning of sampling stations, relative to outfalls, in freshwater versus marine habitats) (Lowell et al., 2003).

The fish survey was designed to measure key endpoints related to age, energy use (growth rate, reproductive investment) and energy storage (condition factor, liver size) (Lowell et al., 2003). An eutrophication response would be expected to result in increased energy use and storage, and this was indicated by some components of the national average response seen in fish (increases in weight at age, condition, liver weight). In contrast, an average decrease in gonad size was also found that, in combination with the increases in the other fish endpoints, was similar to patterns previously observed at some pulp and paper mills prior to the initiation of the EEM program (Munkittrick et al., 1994a). This pattern of decreased gonad size, combined with increased energy use and storage, had been interpreted as a form of metabolic disruption (Munkittrick et al., 2002a) or endocrine disruption (Van Der Kraak et al., 1992), associated with difficulties in producing sufficient sex steroid hormones (McMaster et al., 1996b) . Male and female fish showed similar responses for all endpoints, with significantly greater decreases in gonad weight being observed for female fish ( P < 0.05) (Lowell et al., 2003).

Improvements in effluent treatment and pulping processes had resulted in some recovery from the reductions in gonad size associated with pulp mill effluents in the early 1990s, and it was clear that not all pulp mills were associated with ecologically significant impacts on fish (Munkittrick et al., 2002a). The finding, however, of an average national pattern consistent with early studies on a limited number of pulp and paper mills suggested that there were still potential concerns regarding pulp and paper mill effects on the reproduction of fish via alterations in levels of sex steroid hormones. At the same time, the measured increases in energy use and storage suggested that there was sufficient food for the fish, consistent with the eutrophication responses evident in the invertebrate studies (Lowell et al., 2003).

A special EEM theme issue of the Water Quality Research Journal of Canada outlined some of the key EEM research issues in a peer-reviewed forum in order to increase the circulation and availability of the information developed over the last number of years (Munkittrick et al., 2002b). In addition to an overview of the pulp and paper program, the issue included a variety of papers describing results from the first two EEM cycles including papers describing some of the concerns related to data interpretation and sampling frequency, the development of alternative approaches as well as non-lethal sampling methodologies (Munkittrick et al., 2002b). The National EEM Science Committee developed a variety of research needs based on the experiences of the scientists after the first two cycles of EEM monitoring. In freshwater, the highest research priorities related to fish monitoring were related to developing non-lethal sampling methodologies and developing additional information on home ranges, mobility and exposure of small-bodied forage fish species. As noted after cycle 1, there was also a need to develop additional information on the basic life history of these species, and on how to estimate reproductive performance of multiple spawners using a single sampling event, and the potential sampling designs for multiple spawners using sequential sampling (Munkittrick et al., 2002a).

For fish studies in the marine environment, significant progress was made between Cycles 1 and 2. On the Atlantic coast, it was determined that the small bodied, littoral mummichog (Fundulus heterclitus) could be easily and quickly sampled and provided all of the information required by the Fish Survey. It was present at six of the mills, and at the other mills other small-bodied species such as rock gunnels and cunners were present (Courtenay et al., 2002). For the Pacific coast, small-bodied littoral species would likely play little, if any, role in cycle 3. Unlike the Atlantic coast where many mills discharge effluent into estuaries and shallow water near shore, most B.C. mills discharge effluent at depth and have trapped plumes. For some of these sites, surveys of wild bivalves or snails may be suitable, whereas at other sites caged bivalves may be the best option. Areas requiring continued research in the marine environment included better methods for the collection of data on age, sex and reproduction in bivalves and in other potentially useful sentinels such as the gastropod-molluscs sampled in Quebec during cycle 2 (Courtenay et al., 2002).

Kovacs et al. (2002) conducted studies on the St. François River to assess the biological condition of fish in the vicinity of three pulp and paper mills using three approaches different than what is used for the EEM studies (Table 1). It was there intent to determine whether the use of a characterization of the fish community structure using the Index of Biotic Integrity (IBI) predicted improvements in receiving water quality better than a fish health assessment index (HAI) or the measurement of two biochemical parameters, hepatic mixed function oxygenase activity and plasma steroid levels. Using the IBI approach, they compared collections conducted in 1991 by the Quebec Ministry of the Environment (Richard, 1996), with collections they conducted in 1998. During this time, two of the mills installed secondary effluent treatment systems, while the third mill already had secondary treatment of their effluent. The 1991 study documented reductions in the IBI downstream of the two mills that had only primary treatment of their effluents, but demonstrated increases in IBI downstream of the mill with secondary treatment (Richards, 1996). The 1998 collections demonstrated improvements in the fish communities downstream of the two mills that installed secondary treatment (Kovacs et al., 2002). The IBI downstream of the third mill that had secondary treatment during both collections also showed improvement in 1998, which corresponded to altering the processes somewhat between collections, with corresponding reductions in BOD and TSS (Kovacs et a., 2002). However, the IBI index was still higher downstream of the discharge relative to upstream.

In comparison, the collections for the other two methods (HAI and biochemical parameters) were conducted in September 1995, one month following the installation of secondary treatment at the two upstream mills. For the 1998 collections, the HAI approach demonstrated conflicting results with the two species used at two of the mills. Where tessellated darters (Ethoeostoma olmstedi) were used in 1998, HAI values were actually reduced downstream of the mill discharge, but in comparison, white sucker demonstrated increased HAI values (Kovacs et al., 2002). The biochemical parameters failed to demonstrate any mill related responses in terms of decreased steroid levels or increased MFO activity in 1998 suggesting that either the compounds responsible for these effects were not present or that the receiving environment protected fish from these chemicals. All mills on this system had secondary treatment during all collections for the biochemical comparisons, and other authors have documented recovery in these endpoints following the installation of secondary treatment at other locations (Sandstrom, 1994, 1996; Munkittrick et al., 1997a). Unfortunately the authors failed to compare the three approaches to the EEM sentinel species approach that was also used at these three sites for their cycle 1 and 2 collections. They conclude however that the IBI community approach was capable of demonstrating improvements following the installation of secondary treatment at two sites and process changes at the third location.

Studies initiated in attempts to determine whether effluents from modern mills also caused reproductive alterations in wild fish within the Northern River Basins were completed. Fish health was examined and compared to a number of reproductive endpoints developed to assess endocrine function. As these Northern Rivers were nutrient limited, fish responded to effluent addition (increased nutrients) by showing signs of increased growth and development relative to the upstream reference fish (McMaster et al., 2000b, 2002b). Some signs of altered reproductive function were observed downstream of the mill sites, however additional responses were observed downstream of the municipal sewage treatment plants from the neighbouring communities. For example, longnose sucker collected downstream of the municipal sewage discharge from the city of Grande Prairie, demonstrated alterations in circulating steroid hormone levels and vitellogenin levels relative to the upstream reference fish (McMaster et al., 2002b). The reproductive responses at these three mills were minimal compared to earlier studies examining reproductive function in fish exposed to pulp and paper mill effluents in Jackfish Bay on Lake Superior and in Kapuskasing and Smooth Rock Falls in the Moose River Basin. In fact, in terms of EEM monitored endpoints, fish either showed no responses or a response to nutrient enrichment with increased growth, condition, liver size and gonadal development (McMaster et al., 2002b). Studies were also conducted at these locations examining effluents from these sites for endocrine active compounds (See Characterization of Bioactive Chemicals). Effluents from each of the three mills examined in the wild fish studies contained compounds that bound to both estrogen and androgen receptors as well as to sex steroid binding proteins. Whether the nutrient enrichment response in the fish masked the potential for reproductive effects however was not known.