Turkey Lakes watershed study: description paper

Note: D.S. Jeffries, I.K. Morrison, and J.R.M. Kelso summarized the objectives, site selection, setup, operation, and scientific deliverables for the Turkey Lakes Watershed Study in a paper delivered to the Canadian Hydrological Symposium "Canadian Research Basins: Successes, Failures and Future" (see Proc. Can. Hydrol. Symp., Banff, Alberta, May 9-11, 1988, CHS No. 17, 117-125). Since then, the content of the paper has been twice revised and updated and the latest (1998) version is reproduced below. Revisions to the original manuscript are indicated by italics.

I. Introduction

During the 1970's, science and government became aware of the potential environmental impact of the long range transport of air pollutants (LRTAP). Concern focussed on acid-forming pollutants, particularly SO₂, as several cases of surface water acidification and loss of fish populations were documented in Scandinavia, Canada, and the US. Atmospheric deposition of acids and other contaminants is a system-wide insult; thus, it is appropriate to use whole basin studies to evaluate biogeochemical processes controlling pollutant-ecosystem interactions. Understanding these processes is necessary for both the effective measurement of LRTAP effects and the development of models to confidently predict impacts over greater space and time. This paper describes the development of a study in the Turkey Lakes Watershed (TLW) located in central Ontario (Algoma District), and emphasizes the basin selection process, difficulties encountered, accomplishments, and overall impact in the context of the wider LRTAP research program.

I.1 Basin Selection

Recognizing that there exists wide variations in climate, magnitude of acidic deposition, and terrain characteristics throughout eastern Canada (i.e. that portion of the country affected by acid rain), a series of 5 research basins was selected in which intensive whole-ecosystem monitoring was or could be undertaken. Similarities and differences in responses to LRTAP observed among the basins provide insight into the dominating processes that occur at each location. Two of the basins (Experimental Lakes Area, north-western Ontario near Kenora, and Muskoka-Haliburton, south-central Ontario, centered around Dorset) were established earlier to address the issue of lake eutrophication and already possessed substantial data bases suitable for calculation of ion budgets and assessment of LRTAP effects. The remaining 3 basins were specifically established for LRTAP studies, and included the TLW in central Ontario, north of Sault Ste Marie, Lac Laflamme in southern Quebec, north of Quebec City, and Kejimkujik in southern Nova Scotia. All of them were operating by the early 1980's. They cover the entire range in observed deposition, i.e. from <10 to >30 kg.ha^-1.yr^-1 wet SO₄. The research emphasis in any individual basin tends to reflect its particular physical, chemical, and biological characteristics. Certainly, no single basin does or can address all aspects of atmospheric, aquatic, and terrestrial LRTAP research; however, when taken together, they provide insight into virtually all the questions being pursued during the current national assessment of acid rain effects.

The decision to establish a research basin in the Algoma region of central Ontario was motivated initially by 2 factors: first, a perceived need to fill a gap not covered by the other basins, i.e. investigate LRTAP effects under conditions of significant though not maximum deposition levels in a terrain of only moderate sensitivity; and second, a desire to utilize expertise in a truly integrated watershed study that was already resident in Sault Ste Marie at the Great Lakes Forest Centre (Canadian Forestry Service, Department of Natural Resources) and the Great Lakes Laboratory for Fisheries and Aquatic Science (Department of Fisheries and Oceans). Note that annual bulk SO₄ deposition has ranged from 21 (1987 and 1994) to 40 (1982) kg.ha^-1 between 1982 and 1996 (R.G. Semkin, pers. comm.). Annual deposition was undoubtedly higher before TLW monitoring began (Kelso and Jeffries, 1988), whereas 4 of the 5 lowest annual values have occurred in the most recent years (1993 through 1996). The annual deposition level is therefore somewhat higher than was expected during the original planning of the study; however, this has neither hindered the attainment of objectives nor detracted from the usefulness of results.

A systematic process was initiated to select the field site. Essential basin criteria included: (1) an undisturbed, interconnected, cascading lake chain progressing downstream from an headwater lake, (2) hydrological systems that were amenable to weir construction and accurate flow measurement, (3) geochemical conditions that were regionally representative, (4) location within daily commuting distance of Sault Ste Marie (maximum 50 - 70 km) and having reasonable access, (5) an intact, typical and representative, mature, homogeneous Great Lakes - St Lawrence forest, (6) a gradation in fish population characteristics within the lake chain, (7) self-sustaining fish stocks, and (8) and a salmonid population (brook and/or lake trout) within at least 1 of the lakes. It was assumed that if acceptable basin characteristics were found for fish, the characteristics of lower aquatic biota populations would also be "representative" of other lakes in the region.

By using the simplest logistical criterion (i.e. distance from Sault Ste. Marie), approximately 100 potential candidates were identified on topographic maps. Consideration of existing information (mostly on fish stocks but also other geological and hydrological data) eliminated 70% from further consideration. The remaining basins were visually inspected to determine whether suitable forest conditions existed, and in the end, 8 specific candidates were sampled chemically and biologically to ascertain overall suitability. It turned out that the most restrictive criterion was the combination of basin access and the existence of an intact, mature forest since almost all bush access in the Algoma District has been gained by the construction of logging roads. Also, construction of logging roads usually resulted in substantially increased recreational usage of the lakes and sometimes cottage development; both of these factors reduced the attractiveness of a number of sites. Of the 3 acceptable sites left at the end of the selection process, the Turkey Lakes Watershed (Figure 1) was ultimately chosen as the best overall candidate. It is located approximately 50 km north of Sault Ste. Marie, and, while this was near the outer limit of the distance criterion, it was truly remote and undeveloped, and access was gained at relatively low cost by re-establishing a road previously used to service a nearby but now abandoned fire-tower. The association of the road with the fire-tower rather than logging accounts for the relatively intact nature of the surrounding forest. The entire selection process took 8-9 months to complete.

II. Basin Description

The TLW is 10.5 km² in area and fully contains the drainage area of a headwater chain of 4 lakes (5 distinct basins). It is located on provincial Crown land and is reserved for research purposes through cooperation of the Sault Ste Marie District, Ontario Ministry of Natural Resources. The terrestrial and aquatic resources within the basin are representative of the surrounding region of Algoma, although for Ontario, it does have high relief (290 m) and high annual precipitation (>1200 mm). Bedrock in the TLW is predominantly Precambrian metamorphic basalt (greenstone) with minor occurrences of granite; the overburden is composed of a 2 component glacial till varying in thickness from <1 m at high elevations to 1 - 2 m at lower elevations. There are also limited occurrences of extremely deep till sequences (up to 70 m) within bedrock valleys. The till mineralogy is more felsic than the underlying bedrock showing that this material was likely primarily derived from the large granitic intrusions that occur just north of the basin. Also, the till contains 0 - 2% CaCO₃ with higher concentrations generally occurring deeper in the profile and at lower elevations. This small but easily weathered component of the overburden is the primary factor contributing to the "moderate" sensitivity of the terrain and is a feature common to much of northern Ontario (Jeffries et al., 1986a).

The rather angular drainage pattern (Figure 1) arises from structural control exerted by 2 intersecting regional fault systems. There have been as many as 20 fully instrumented streamflow gauging stations, most having artificial flow control structures. Most gauging stations are situated on first order headwater streams; the remainder are at lake outflows or further downstream on the major stream channel (Norberg Creek). Hydrological monitoring (like all other chemical, meteorological, and deposition monitoring) is conducted on a year-round basis.

The chain of lakes range from 5.8 ha (southern basin of Batchawana Lake, see Figure 1) to 52.0 ha (Turkey Lake) in surface area, and from 2.2 m (Wishart Lake) to 12.2 m (Turkey Lake) in mean depth. The relatively high precipitation causes the lakes to flush fairly rapidly; water renewal times range from 0.15 (Wishart L.) to 1.3 yr (Batchawana Lake North). All of the lakes are dimictic with the exception of Wishart L. which experiences recurrent wind-induced mixing due to its shallowness. Dissolved oxygen depletion occurs in undisturbed bottom waters of all lakes except Turkey Lake; the hypolimnetic oxygen pool present in this the deepest lake of the chain is apparently large enough to prevent development of anoxic conditions.

There is a gradient in the major ion composition of lakes within the TLW, the most dilute waters occurring at high elevations. Calcium levels increase from 55 to 138 mmol.L^-1 down the chain. Sulphate is the dominant anion in Batchawana L. while alkalinity dominates in the lower lakes. Phosphorus is the limiting nutrient in these lakes. Nitrate nitrogen exhibits relatively high mean levels (7.9-16.4 mmol.L^-1) and fairly pronounced seasonal cycles because the terrestrial basin exhibits low utilization of this nitrogen species. Seasonal and episodic variation in several surface water variables can be large.

At the beginning of the study, Batchawana Lake contained no fish. However, trout were later introduced to the southern basin and their population status subsequently monitored. Fish communities in the lower 3 lakes (composed of 3-11 species) are typical of the Algoma region. The distribution of benthic organisms is primarily a function of lake depth and the presence/absence of fish rather than variations in water chemistry. Zooplankton species composition is similar across all lakes, and cyanophytes are the dominant algae throughout.

The forest is an uneven-aged, mature-to-overmature, old growth tolerant hardwood stand. The principal tree species are maples (90%), with lesser amounts of other hardwoods (9%, usually yellow birch) and various conifers (1%). Several terrestrial sub-basins draining into Norberg Creek below the outflow of Turkey L. were logged in 1997 using a variety of forest harvesting techniques for the purpose of evaluating their differing ecological impacts. This forest harvesting project is expected to continue until at least 2010. Forest production is typical of that observed at this northerly latitude (47EN). Foliar bioelement concentrations are generally similar to those observed at Hubbard Brook, New Hampshire. Further detail on the physical, chemical, and biological characteristics of the TLW can be found in Jeffries et al. (1988).

III. Discussion

III.1 Research Topics

The TLWS was intended from the outset to be multi-disciplinary in nature, and the breadth of research conducted in the basin reflects this priority. Primary research responsibilities fall on several government agencies including the Great Lakes Forest Centre (GLFC) of Natural Resources Canada, the National Water Research Institute (NWRI) of Environment Canada, the Great Lakes Laboratory for Fisheries and Aquatic Science (GLLFAS) of Fisheries and Oceans Canada, the Atmospheric Environment Service (AES) of Environment Canada, the Canadian Wildlife Service (CWS) of Environment Canada, and the Ontario Ministry of Natural Resources (OMNR). Various university researchers have also been involved (Queen's, Guelph, McMaster, Laurentian, Waterloo, Laurier, New Brunswick, McGill, Toronto, Western). In order to coordinate activities among such a diverse group and maintain overall control of the study, scientific direction is provided by a Steering Committee composed of senior scientists from the participating agencies.

The TLW station for measurement of atmospheric deposition has been designated an official site in the Canadian Air and Precipitation Monitoring Network (AES). Both wet and dry deposition is measured (Sirois and Vet, 1988). In addition, meteorology and bulk deposition are monitored by NWRI (Semkin and Jeffries, 1986a) and bulk deposition, throughfall, and stemflow have been monitored by GLFC (Foster and Nicolson, 1983). Extensive studies have been conducted on snowpack storage and release of pollutants, and the associated effects on basin hydrochemistry (Semkin and Jeffries, 1986b; English et al., 1986). Finally, measurement of the atmospheric deposition of synthetic organic (toxic) compounds has been conducted (Johnson et al., 1988; Strachan and Huneault, 1984) and plans are being developed to evaluate terrestrial retention of these compounds as a mechanism regulating their input to aquatic ecosystems.

Extensive studies of water quality development have been conducted in the TLW primarily by GLFC and NWRI. These include description of streamflow generation and the factors affecting stream composition (Semkin et al., 1984; Nicolson et al., 1987), groundwater hydrogeochemistry (Bottomley et al., 1986; Foster et al., 1986), and lake chemistry (Jeffies et al., 1986b). Episodic changes in surface water composition were addressed by Jeffries and Semkin (1983). Paleolimnolgical analyses of lake sediment cores have been conducted using diatom (Delorme et al., 1986), metal profiles (Johnson et al., 1986), and invertebrate remains (Johnson and McNeil, 1988).

Aquatic biology research is primarily conducted by GLLFAS and has encompassed surveys of biological communities (Kelso et al., 1982), evaluations of phytoplankton production (Collins et al., 1983), benthic production (Dermott, 1988), fish population production (Kelso, 1985), and factors affecting fish production (Kelso, 1988). Habitat manipulation studies have been proposed by GLLFAS to evaluate the impact of altered habitat on fish biomass, production, and community structure. It will involve modification of some of the littoral area of Wishart and/or Little Turkey Lakes (e.g. removal of woody debris) while using Batchawana L. as the "no-manipulation" control. Littoral zone modification will not occur before 2000.

Terrestrial studies have included extensive evaluations of soil physical and chemical characteristics (Cowell and Wickware, 1983), laboratory studies of soil responses to acidic deposition (Hern et al., 1985; Hay et al., 1985), element cycling in the forest ecosystem (Foster et al., 1983; Foster and Nicolson, 1986), quantification of phytomass element concentrations (Morrison, 1983; Morrison and Hogan, 1986), and estimation of timber standing stock and production. In 1997, a forest harvesting project was initiated involving logging some of the terrestrial sub-basins below Turkey L. by 3 different techniques (clear-cut, selection and shelterwood cutting). The project is intended to evaluate the impact of the differing techniques on soils; stand structure, function and net production; sub-basin water yield and water quality; and overall ecosystem biodiversity. Model development is also an important component.

Finally, there has been extensive work to develop and validate simulation models for predicting stream hydrology and chemistry (Bobba et al., 1986; Lam et al., 1986). They have proven to be quite transportable having successfully simulated flow and chemistry from other watersheds. The models along with terrestrial watershed (Foster et al., 1986) and lake mass budget studies (Jeffries et al., 1986b) have also served to integrate many of the individual results into a broader picture. The overall integration of component studies still requires much future effort however. It should be noted that due to space limitations, this is an incomplete list of topics researched within the TLWS. Many other papers deal with such areas as stable isotope investigations, microbiology, additional lake sediment studies, geology/geochemistry, chemical and biological trends, nitrogen-based acidification, wildlife, forest classification, etc.

III.2 Difficulties

The most significant difficulties encountered during the life of the TLWS have been logistical, arising from the basin's remote location. Most of the problems have been solved, but at significant and continuing costs. The most important of these are fuelling and maintaining diesel generators to supply the electrical needs of the atmospheric monitoring station and the small field laboratory, and road maintenance (particularly snow ploughing) to permit daily year-round access to the site. Accessibility during the peak of spring snowmelt can be very difficult but is nevertheless essential since there has been great emphasis placed on defining effects of the snowpack release of pollutants; thus, a small camp of "bunk house" trailers has been established to accommodate 6 - 12 field technicians for this short period of time. The distance of the basin from Sault Ste Marie also costs substantial time, i.e. that unproductive commuting time spent by field personnel. However, we believe that the quality of the ecosystem available for study in the TLW outweighs all of these problems.

Another potential difficulty is the multiple agency structure of the overall study. First, sharing of major logistical costs among the 3 principal participating agencies (NWRI, GLFC, and GLLFAS) has left the study vulnerable to cost-cutting by any one of them. While this has not been a major problem to date, it continues to be worrisome since the agencies are meeting these obligations using "soft" resources. Second, shifts in research priorities by different agencies could threaten the cohesive nature of the TLWS. The active role taken by the Steering Committee does much to promote inter-agency communication and maintain the harmonious nature of the independent, yet interdependent studies being conducted in the basin. As of 1998, the cooperative cost sharing approach has permitted collection of an unbroken monitoring record, but shifts in priorities of the primary partners may require a re-evaluation of existing operating relationships within the near future.

III.3 Accomplishments, Impact, Availability of Results

Acid rain has been a high priority research issue for all of the agencies involved in the study. From its inception therefore, the TLWS has been under pressure to produce information that is of immediate value in assessing local and national acidification effects. The study has been very successful in this regard as reflected in publication of well over 250 reports and/or papers covering all research components: atmospheric deposition, water quality development, geochemical process definition, aquatic and terrestrial biology, episodic acidification, simulation modelling, etc (see the TLW Publication List). Results were extensively cited in the most recent national assessment of acid rain effects on the Canadian environment (RMCC, 1986; note that the 1986 Assessment was superseded by RMCC, 1990; and later by Environment Canada, 1997; Jeffries, 1997; and Hall et al., 1997; all contain numerous TLW citations).

Three particular products attest to the scientific impact of the TLWS. First, at a time when the TLWS had been in existence for only 5 years, participating scientists presented and published 10 papers at the last major international acid rain symposium, the so-called Muskoka Conference in 1985 (Muskoka Conf. Proc. appear in Vol. 30 and 31 of Wat. Air Soil Pollut.). Second, a more important and thorough presentation of the wide-ranging research conducted in the basin appeared as a special volume of Can. J. Fish. Aquat. Sci. (21 papers in Vol. 45, Suppl. 1). A second special volume is now in preparation. Third, the Steering Committee decided early on that an inclusive reference list be maintained for all publications related to the TLWS. The Reference List contains a short summary of each entry and is available to anyone (see "Publications" in this web site). It has served as a useful mechanism for disseminating knowledge of study findings among the greater scientific community. As of October, 1998, the Reference List contains 251 entries.

Finally, if confronted with the same research need again, we feel confident that the research basin approach would still be adopted. Evidence of this is the continued operation of the TLWS, the current evolution of research activities to address the problems of NO_x, metals, toxic contaminants, UV-b and climate change, and a new initiative to designate the site as a member of an international long-term integrated monitoring network (the TLW is presently the only Canadian site included in the UNECE "Integrated Monitoring" program).

IV. References

Bobba, A.G., Lam, D.C.L., Jeffries, D.S., Bottomley, D., Charette, J-Y., Dillon, P.J., and Logan, L. 1986. Modelling the hydrological regime in acidified watersheds. Wat. Air Soil Pollut. 31:155-164.

Bottomley, D.J., Craig, D., and Johnston, L.M. 1986. Oxygen-18 studies of snowmelt runoff in a small Precambrian Shield watershed: implications for streamwater acidification in acid-sensitive terrain. J. Hydrol. 88:213-234.

Collins, R.H., Love, R.J., Kelso, J.R.M., Lipsit, J.H., and Moore, J.E. 1983. Phytoplankton production as estimated by the ¹⁴C technique, and populations contributing to production in the Turkey Lakes Watershed. Cdn. Tech. Rep. Fish. Aquat. Sci., No. 1191, 23p.

Cowell, D.W. and Wickware G.M. 1983. Preliminary analyses of soil chemical and physical properties, Turkey Lakes Watershed, Algoma, Ontario. Turkey Lakes Watershed Unpub. Rep. No. 83-08, 25p.

Delorme, L.D., Duthie, H.C., Esterby, S.R., Smith, S., and Harper, N.S. 1986. Prehistoric inferred pH changes in Batchawana Lake, Ontario from sedimentary diatom assemblages. Arch. Hydrobiol. 108:1-22.

Dermott, R.M. 1988. Distribution of benthic invertebrates in the Turkey Lakes Watershed. Can. J. Fish. Aquat. Sci. 45(Suppl1):108-114.

English, M.C., Jeffries, D.S., Foster, N.W., and Semkin, R.G. 1986. A preliminary assessment of the chemical and hydrological interaction of acidic snowmelt water with the terrestrial portion of a Canadian Shield catchment. Wat. Air Soil Pollut. 31:27-34.

Environment Canada. 1997. 1997 Canadian Acid Rain Assessment, Volume Two: Atmospheric Sciences, Ottawa, Ontario.

Foster, N.W. and Nicolson J.A. 1983. Ion transfer through a tolerant hardwood canopy, Turkey Lakes Watershed, Ontario. Proc. Conf. Acid Rain and For. Resourc., Quebec City.

Foster, N.W., Nicolson, J.A., and Morrison, I.K. 1983. Acid deposition and element cycling in eastern North American forests. Proc. Conf. Acid Rain and For. Resourc., Quebec City.

Foster, N.W. and Nicolson, J.A. 1986. Trace elements in the hydrologic cycle of a tolerant hardwood forest ecosystem. Wat. Air Soil Pollut. 31:501-508.

Foster, N.W., Morrison, I.K., and Nicolson, J.A. 1986. Acidic deposition and ion leaching from podzolic soil under a hardwood forest. Wat. Air Soil Pollut. 31:879-889.

Hall, P., Bowers, W., Hirvonen, H., Hogan, G., Foster, N., Morrison, I., Percy, K., Cox, R. and Arp, P. 1997. 1997 Canadian Acid Rain Assessment, Volume Four: The Effects on Canada's Forests, Ottawa, Ontario.

Hay, G.W., James, J.H., and Van Loon, G.W. 1985. Solubilization effects of simulated acid rain on the organic matter of forest soils: preliminary results. Soil Sci. 139:422-430.

Hern, J.A., Rutherford, G.K., and Van Loon, G.W. 1985. Chemical and pedogenic effects of simulated acid precipitation on 2 eastern Canadian soils. Can. J. For. Res. 15:839-847.

Jeffries, D.S. 1997. 1997 Canadian Acid Rain Assessment, Volume Three: The Effects on Canada's Lakes, Rivers and Wetlands, Ottawa, Ontario.

Jeffries, D.S. and Semkin, R.G. 1983. Changes in snowpack, stream, and lake chemistry during snowmelt in the Turkey Lakes Watershed. VDI-Berichte Nr. 500:377-386.

Jeffries, D.S., Wales, D.L., Kelso, J.R.M., and Linthurst R.A. 1986a. Regional chemical characteristics of lakes in North America: Part I - Eastern Canada. Wat. Air Soil Pollut. 31:551-567.

Jeffries, D.S., Semkin, R.G., Neureuther, Seymour, M.D. and Nicolson, J.A. 1986b. Influence of atmospheric deposition on lake mass balances in the Turkey Lakes Watershed, central Ontario. Wat. Air Soil Pollut. 30:1033-1044.

Jeffries, D.S., Kelso, J.R.M., and Morrison, I.K. 1988. Physical, chemical, and biological characteristics of the Turkey Lakes Watershed, central Ontario, Canada. Can. J. Fish. Aquat. Sci. 45(Suppl 1):3-12.

Johnson, M.G., Culp, L.R., and George, S.E. 1986. Temporal and spatial trends in metal loadings to sediments of the Turkey Lakes, Ontario. Can. J. Fish. Aquat. Sci. 43:754-762.

Johnson, M.G., Kelso, J.R.M., and George, S.E. 1988. Loadings of organochlorine contaminants and trace elements to two Ontario lake systems and their concentrations in fish. Can. J. Fish. Aquat. Sci. 45(Suppl 1):170-178.

Johnson, M.G. and McNeil, O.C. 1988. Fossil midge associations in relation to trophic and acidic state of the Turkey Lakes. Can. J. Fish. Aquat. Sci. 45(Suppl 1):136-144.

Kelso, J.R.M. 1985. Standing stock and production of fish in a cascading lake system on the Canadian Shield. Can. J. Fish. Aquat. Sci. 42:1315-1320.

Kelso, J.R.M., Love, R.J., Lipsit, J.H., and Dermott, R. 1982. Chemical and biological status of headwater lakes in the Sault Ste Marie District, Ontario. In: Acid Precipitation, Effects On Ecological Systems, F.M. D'itri (ed), Ann Arbour Science Publishers, Ann Arbour, MI, 165-207.

Kelso, J.R.M. 1988. Fish community structure and production in a watershed in north central Ontario. Can. J. Fish. Aquat. Sci. 45(Suppl 1):115-120.

Kelso, J.R. M. and Jeffries, D.S. 1988. Response of headwater lakes to varying atmospheric deposition in north-central Ontario, 1979 - 1985. Can J. Fish. Aquat. Sci. 45:1905-1911.

Lam, D.C.L., Boregowda, S., Bobba, A.G., Jeffries, D.S., Patry, G.G. 1986. Interfacing hydrological and hydrogeochemical models for simulating streamwater chemistry in the Turkey Lakes Watershed, Canada. Wat. Air Soil Pollut. 31:149-154.

Morrison, I.K. 1983. Biomass and macroelements in a tolerant hardwood stand, Turkey Lakes Watershed, Ontario. Proc. Conf. Acid Rain and For. Resourc., Quebec City.

Morrison, I.K. and Hogan, G.D. 1986. Trace element distribution within the tree phytomass and forest floor of a tolerant hardwood stand, Algoma, Ontario. Wat. Air Soil Pollut. 31:493-500.

Nicolson, J.A., Craig, D., and Foster, N.W. 1987. Precipitation, surface, and subsurface water chemistry in a tolerant hardwood forest basin. Proc. Int. Symp. Acidif. and Wat. Pathways, Bolkesjo, Norway, 91-100.

RMCC. 1990. The 1990 Canadian Transport of Air Pollutants and Acid Deposition Assessment Report. Federal/Provincial Research and Monitoring Coordinating Committee, Ottawa. Part 3: Atmospheric Sciences; Part 4: Aquatic Effects; Part 5: Terrestrial Effects.

Semkin, R.G., Jeffries, D.S., and Neureuther, R. 1984. Relationships between hydrological conditions and the ionic composition of streamwaters in the Turkey Lakes Watershed. Proc. Can. Hydrol. Symp., Quebec City, 109-122.

Semkin, R.G. and Jeffries, D.S. 1986a. Bulk deposition of ions in the Turkey Lakes Watershed. Wat. Pollut. Res. J. Can. 21:474-585.

Semkin, R.G. and Jeffries, D.S. 1986b. Storage and release of major ionic components from the snowpack in the Turkey Lakes Watershed. Wat. Air Soil Pollut. 31:155-164.

Sirois, A. and Vet, R.G. 1988. Atmospheric deposition of acid-related substances to the Turkey Lakes Watershed. Can. J. Fish. Aquat. Sci. 45(Suppl 1):26-37.

Strachan, W.M.J. and Huneault, H. 1984. Automated rain sampler for trace organic substances. Environ. Sci. Technol. 18:127-130.