20: Additional Problems of Surface Waters

Causes of Eutrophication

Most lakes in Canada are geologically “young” because they occur on landscapes that were released from glacial ice only about 8-12-thousand years ago (depending on the region). Many lakes are relatively deep, having had little time to accumulate much sediment in their basins. They also tend to have low rates of nutrient input. Over the millennia, however, oligotrophic lakes of this sort gradually increase in productivity as they accumulate nutrients. They also become shallower due to sedimentation, which results in increased rates of nutrient cycling. The slowly increasing productivity of many lakes over time is a natural expression of the eutrophication as a longer-term process.

Surface waters may also be naturally eutrophic if they occur in a watershed with fertile soil. This is the case of many lakes and ponds in the prairie region, where there is an abundance of shallow waterbodies that recycle their nutrients quickly.

Anthropogenic (cultural) eutrophication involves more rapid increases in aquatic productivity. This is most often caused by nutrient loading associated with sewage dumping or runoff contaminated by agricultural fertilizer. Wherever humans live in dense populations or engage in intensive agriculture, there are large inputs of nutrients into lakes and rivers. In coastal areas, estuaries and shallow near-shore areas may also be affected by anthropogenic nutrient inputs and eutrophication.

A theory called the Principle of Limiting Factors states that certain ecological processes are controlled by whichever environmental factor is present in the least supply relative to demand. According to this idea, the primary production of waterbodies is limited by the nutrient that is present in least supply relative to its demand (assuming that light, temperature, and oxygen supply are all adequate). Research has shown that, in almost all fresh waters, primary production is limited by the availability of phosphorus, occurring as the phosphate ion (PO₄^3–. In contrast, marine waters are usually limited by the availability of inorganic nitrogen, particularly in the form of nitrate (NO_3-).

Phosphorus as the Limiting Nutrient

During the 1960s and early 1970s, eutrophication was a topic of intense environmental controversy, part of which focused on the question of which nutrients were the limiting factors to primary production in lakes. Many scientists believed that phosphorus was most important in this regard. Others, however, suggested that dissolved nitrogen, in the form of nitrate or ammonium, was the critical limiting nutrient in many fresh waters. Plants and algae have relatively large demands for all of those nutrients, which typically occur in low concentrations in water.

It was also suggested that dissolved inorganic carbon (as bicarbonate, HCO_3–) could be a limiting factor. Inorganic carbon is needed in large amounts by autotrophs, but in aquatic systems HCO_3– is replenished mainly by the diffusion of atmospheric CO₂ into the surface water, which is a slow process.

Eventually, research convincingly demonstrated that phosphorus is the key nutrient that limits primary production in most fresh waters. Consequently, controlling the rate of phosphorus supply is now known to be a crucial action if eutrophication is to be avoided.

The vital role of phosphorus is suggested in Table 20.1, which shows its typical concentration in fresh water (an index of supply) and its concentration in plants (an index of demand). Because the ratio of demand to supply for phosphorus is considerably larger than for other nutrients, it is a likely candidate to be the primary limiting factor for the productivity of algae and plants. The data also suggest that the next-most important nutrient is inorganic nitrogen, in the form of nitrate or ammonium.

Other studies confirm that phosphorus is the usual controlling nutrient for eutrophication. Perhaps the most convincing data come from a famous series of experiments conducted in the Experimental Lakes Area (ELA; see Canadian Focus 20.1). Most of the research at the ELA involved the addition of nutrients at various rates and in different combinations to selected lakes, followed by monitoring of the ecological responses. The most important whole-lake experiments were the following (Schindler, 1978, 1990; Levine and Schindler, 1989):

• For two years, Lake 304 was fertilized with phosphorus, nitrogen, and carbon. It responded by becoming eutrophic. After the addition of phosphorus was stopped, the lake returned to its original oligotrophic condition, even though nitrogen and carbon were still being added.
• The two basins of Lake 226, an hourglass-shaped lake, were isolated with a vinyl curtain. One basin was fertilized with carbon, nitrogen, and phosphorus in a weight ratio of 10:5:1, while the other received only carbon and nitrogen at 10:5. Only the basin receiving phosphorus developed algal blooms. After five years, the nutrient additions were stopped, and the original oligotrophic condition returned within just one year. A similarly rapid recovery was observed in another experiment, involving Lake 303, after fertilization with P was stopped.
• Lake 302N was fertilized with phosphorus, nitrogen, and carbon. However, the nutrients were injected directly into deep water during the summer, at a time when lakes develop a thermal stratification (see In Detail 20.1). Because the nutrients were injected deeply, primary production in the surface water was not affected and eutrophication did not occur. Research at the ELA also showed that when phosphorus was added to lakes, the supply of nitrogen and carbon already present was capable of supporting phosphorus-induced eutrophication. In part, this occurred because eutrophication induced an increased rate of fixation of atmospheric dinitrogen (N2) by blue-green bacteria and a faster diffusion of atmospheric CO₂ into the lakewater.

Eutrophication also causes pronounced changes to occur in the species composition, relative abundance, and biomass of phytoplankton, which are the most important primary producers in the ELA lakes. Changes in the phytoplankton resulted in effects on organisms at higher levels of the food web, such as zooplankton and fish. For example, the productivity of whitefish (Coregonus clupeaformis) in hourglass-shaped Lake 226 was greater in the eutrophied basin than in the oligotrophic one. This occurred in response to the greater abundance of their prey of zooplankton and aquatic insects, which itself a trophic response to increased algae productivity.

In Detail 20.1. Lake Stratification
During most of the year, lakes have a rather uniform distribution of temperature throughout their depth. This allows their bottom and surface waters to mix easily under the influence of strong winds. During the summer, however, lakes often develop a persistent stratified condition. This is characterized by a surface layer of relatively warm water, several metres thick in most lakes (but thicker in some large lakes), lying above deeper, cooler water. Because the density of warm water is less than that of cool water, the two layers remain discrete and do not mix.

The relatively warm, upper water of a stratified lake is known as the epilimnion, while the cooler, deeper water is the hypolimnion. These layers are separated by a narrow zone of rapid change, known as the thermocline. During the autumn, as the epilimnion cools, the stratification diminishes and the two layers become mixed by strong winds.

During stratification, oxygen and other dissolved substances can enter the hypolimnion, but mainly by diffusion across the thermocline. Because diffusion is a slow process, hypolimnetic oxygen becomes depleted in stratified lakes whose deeper water receives large inputs of organic material. The organic input may be associated with sewage, agricultural runoff, or algal biomass sinking from the epilimnion. The development of anoxic (no oxygen) conditions represents an important degradation of water quality because fish and most other animals cannot live in such an environment.

Sometimes, density gradients associated with dissolved salts may also cause lakes to become stratified. In these cases, a surface layer of fresh water sits on top of more saline, deeper water, with the layers separated by a steep chemical and density gradient known as a halocline.

Canadian Focus 20.1. The Experimental Lakes Area
Some of the most famous Canadian research in ecology and environmental science has been conducted in the Experimental Lakes Area (ELA). Research at the ELA, funded mainly by the Federal Department of Fisheries and Oceans (DFO), began during the late 1960s. The first two decades of work are closely identified with David Schindler, but hundreds of other Canadian and international scientists from governments and universities have been involved in research in the area. Several of the leading scientists working at the ELA, notably Schindler, have won international awards for their outstanding contributions.

The ELA is located in a remote region of northwestern Ontario near Kenora. The area contains many lakes and wetlands. The watersheds are largely forested, with thin soil that overlies hard quartzitic bedrock of the Canadian Shield. The ELA is not a protected area, so forestry is pursued, as are mining exploration and tourism. However, there is an understanding among the various interests in the ELA that some of the lakes and their watersheds are being used for research and should not be disturbed.

Image 20.1. Lake 226 in the Experimental Lakes Area was divided into two separate basins with a heavy vinyl curtain. The upper basin in the photograph was fertilized with phosphorus, nitrogen, and carbon, while the lower one received nitrogen and carbon. Only the basin that received phosphorus became eutrophic and developed blooms of phytoplankton, which are discernible as a whitish hue in the photo. Source: Michael Paterson.

One of the most important experimental procedures used at the ELA has been the controlled perturbation of entire lakes or wetlands. This is done to investigate the ecological effects of certain anthropogenic stressors, such as eutrophication, acidification, metal pollution, and flooding of wetlands. These are known as whole-lake experiments (and some as whole-wetland experiments). The experimental procedure includes an initial study of the lake or wetland for several years to determine the baseline conditions. The system is then perturbed in some way, for example, by causing it to become eutrophic by adding nutrients. The experimental lakes are monitored for a range of ecological responses, such as changes in the abundance and productivity of species and communities, in nutrient cycling, and in chemical and physical factors.

The experimental lakes are paired with reference (non-perturbed) ones, which provide information on natural changes that are unrelated to the manipulation. Because the reference lakes are monitored for a long time, they also provide extremely useful information relevant to changes in the ambient environment, such as climate warming.

Unfortunately, research at the ELA has been periodically threatened by cutbacks in funding, and the future of this world-class facility and its programs has been subjected to bouts of uncertainty. Throughout the history of the ELA, the DFO provided most of the funding to maintain its research infrastructure. Most of the experiments were also funded by DFO and conducted by its personnel or by university scientists working with them. However, during the early 1990s, DFO focused its activities more so on marine issues, and for several years it looked as if cutbacks might result in the ELA facility closing down. In fact, despite an international outcry from aquatic scientists and environmentalists, the ELA was essentially closed for about a year. During that time several buildings were taken down and extremely important research projects were curtailed or abandoned. Fortunately, however, the provinces of Ontario and Manitoba provided funding that has allowed the International Institute for Sustainable Development to manage and operate the ELA facility, so this vital location for whole-ecosystem science is now operating again to support world-class ecological research.

Sources of Nutrient Loading

In North America, eutrophication was most severe as an environmental problem during the 1960s and 1970s. At that time, the average discharge of phosphorus to inland waters was about 2 kg/person•year. About 84% of the phosphorus loading was associated with the dumping of municipal sewage, and the rest was due to agricultural fertilizer and sewage (from livestock). In addition, the nitrogen discharge at the time was about 12.5 kg/person•year, of which 36% was from municipal and 64% from agricultural sources.

Because scientists have convincingly demonstrated that phosphorus is the primary limiting nutrient for eutrophication in fresh water, control strategies have focused on reducing the input of that nutrient to surface waters.

Sewage Treatment

The production of livestock manure in Canada is about 181 × 10⁶ t/y, with a phosphorus content of 0.30 × 10⁶ t/y and nitrogen content of 1.10 × 10⁶ t/y (in 2006; Statistics Canada, 2008). Not much of the livestock manure is treated before being released to the environment – it is mostly disposed onto fields or into a nearby waterbody. The production of human sewage is about 10 × 10⁹ t/y (most of this is water, equivalent to 10 × 10¹² L/y; because municipal sewage is flushed away in toilets, it is much more diluted by water than is livestock manure; Environment Canada, 2010). Most of the human sewage in Canada is treated before it is released to the environment (see Chapter 25 for notable exceptions).

In most places, the principal objectives of sewage treatment are to reduce inputs of pathogenic microorganisms and oxygen-consuming organic matter to receiving waters. However, in places where surface waters are vulnerable to eutrophication, sewage may also be treated to reduce the quantity of phosphorus in the effluent.

All sizeable towns and cities in Canada have facilities to collect the sewage effluent from homes, businesses, institutions, and factories (Chapter 25). This infrastructure consists of complex webs of underground sewage pipes and other collection devices. Some municipalities have separate systems to collect domestic and industrial wastes, because the latter often contains toxic and hazardous chemicals that should be treated separately. Some municipalities also have a separate system to handle the large volumes of storm flows, which result from the runoff of rain and snow meltwater. Eventually, these large quantities of wastewater must be discharged into the ambient environment, usually into a nearby lake, river, or ocean. Wherever possible, it is highly desirable – and environmentally responsible – to treat the wastewater to reduce its pollutant load before discharging it into an aquatic ecosystem.

Regrettably, however, some municipalities in Canada continue to dump their raw, untreated sewage into a nearby aquatic environment. This practice is more common for cities and towns that are located beside an ocean, because well-flushed marine environments have a huge capacity for diluting and biodegrading organic pollutants. Cities that dump raw sewage into the ocean include Saint John and St. John’s on the Atlantic coast, and Victoria on the Pacific coast. Consequently, some of their coastal habitats have become degraded by the aesthetic, hygiene-related, and ecological damage associated with the dumping of untreated sewage. Although the worst damage is restricted to the vicinity of the sewage outfalls, it is still an important problem that should be responsibly addressed by constructing sewage-treatment facilities.

Compared with many oceanic environments, inland waters such as lakes and rivers have a much smaller capacity for diluting and biodegrading sewage waste. Consequently, most municipalities located beside an inland waterbody treat their sewage before discharging the effluent. Sewage treatment can, however, vary greatly in degree and in the technology used, as we examine below (Freedman, 1995; Sierra Legal Defence, 2004, 2006).

• Primary sewage treatment is relatively simple. It usually involves the screening of raw sewage to remove larger materials, and then allowing the remainder to settle to reduce the amount of suspended organic matter. The resulting effluent is then discharged into the environment, although it may first be treated with a disinfectant (usually a chlorine compound) to kill pathogens, especially bacteria. Primary treatment typically removes 40-60% of the suspended solids of raw sewage and 5-15% of the phosphorus, while reducing the biological oxygen demand (BOD) by 25-40%. (BOD is the capacity of the organic material in wastewater to consume oxygen during decomposition.) More advanced primary systems can reduce the suspended solids by 90% and the BOD by 50%. In addition, they can reduce fecal coliforms by 45-55%.
• Secondary sewage treatment may be applied to the effluent of primary treatment, mostly to further reduce the BOD. Secondary treatment usually involves the use of a biological technology that enhances aerobic decomposition of organic waste by stimulating the microbial community in an engineered environment. Two such biotechnologies in common use are (1) activated sludge, which involves a vigorous aeration of sewage water to enhance the decomposition of its organic content, and (2) trickling filters, in which the wastewater passes slowly through a complex physical matrix that supports a large population of microorganisms. These biotechnologies, along with primary treatment, produce large quantities of a humus-like product known as sludge, which can be composted and then spread onto agricultural land as an organic-rich conditioner. Sludge may also be less-usefully incinerated or dumped in a landfill (see Chapter 18). Primary and secondary treatments together remove 30-50% of the phosphorus from sewage, and reduce the BOD and suspended solids by 85-90% and coliforms by 90-99%.
• Tertiary sewage treatment includes processes to remove most of the remaining dissolved nutrients from the effluent. Phosphorus removal may be achieved by adding aluminum, iron, calcium, or other chemicals that develop insoluble precipitates with phosphate, which then settle out of the water, removing 90% or more of the phosphorus. Other tertiary processes may be used to remove ammonium and nitrate.
• Artificial wetlands are sometimes constructed to provide advanced treatment of sewage. The wetlands are engineered to develop a highly productive ecosystem, in which vigorous microbial activity decomposes organic waste while algae and macrophytes decrease nutrient concentrations in the water. Most sewage-treatment wetlands are constructed outdoors, but some are developed inside of a greenhouse, which allows the system to work during the winter. The efficiency of these systems depends on climate, the rate of flow-through of the sewage, and the nature of the engineered wetland. Typically, artificial wetlands remove up to 30% of the phosphorus from raw sewage, while reducing BOD by as much as 90%.

Tertiary treatment to reduce phosphorus in municipal effluents requires expensive investments in technology and operating costs. Consequently, this practice is pursued only under certain conditions. Tertiary treatment is mostly used by communities located beside rivers and around the Great Lakes. Moreover, because the Great Lakes are affected by effluents originating from sources in both Canada and the United States, bilateral agreements have been negotiated concerning the loading of phosphorus and other pollutants. To meet the target loadings, municipalities in both countries must use tertiary systems to achieve a high degree of phosphorus removal from sewage.

Elsewhere in Canada, less attention is generally paid to the removal of phosphorus from municipal effluents. Although municipalities may treat their sewage, only primary or secondary systems are generally used, mostly to reduce the abundance of pathogens and to lower BOD in the effluent (Chapter 25).

As previously noted, agricultural livestock produce enormous amounts of fecal materials. However, their manure is rarely treated before it is disposed into the environment. Treatment facilities for agricultural sewage are considered too expensive and are not often required by regulators. This happens even though some intensive rearing facilities, such as agro-industrial feedlots and factory farms, may produce huge amounts of concentrated manure, equivalent to the sewage of a small city.

Eutrophication of an Arctic Lake

Meretta and Char Lakes are two small lakes that are located on Cornwallis Island in northern Canada. Because of the severe climate in the High Arctic, tundra lakes are relatively simple ecosystems. Moreover, nutrient inputs are sparse and cycling is slow, so tundra lakes are naturally oligotrophic. Char Lake is a typical oligotrophic, polar lake, with extremely clear water and low rates of productivity of algae, zooplankton, and fish. Meretta Lake, in contrast, received sewage from a small community. Because of the nutrient input, it became moderately eutrophic (Schindler et al., 1974).

The sewage dumping resulted in phosphorus loading about 13 times larger in Meretta than in Char Lake, while the nitrogen input was 19 times higher. Consequently, during the growing season, Meretta Lake developed a phytoplankton biomass averaging 12-times greater than in Char, and up to 40-times higher during the summer algal bloom. In the winter, these lakes are covered with ice, which restricts the rate at which atmospheric oxygen can enter the water. During this time, the decomposition of organic material, mostly sewage waste but also algal biomass, exerts a great demand for oxygen in the bottom water of Meretta Lake, resulting in anoxic conditions. The oxygen depletion causes severe stress to aquatic animals and impairs the reproduction of Arctic char (Salvelinus alpinus), a type of trout.

This case demonstrates that even polar lakes, which are relatively simple ecosystems because of their severe climatic environment, can exhibit a strong eutrophication response to fertilization with limiting nutrients.

Lake Erie: Eutrophication and Other Stressors

The Great Lakes are one of the world’s outstanding fresh water systems, holding about one-fifth of global surface fresh water (excluding glaciers). Lake Superior sits at the top of this chain, with Lakes Michigan, Huron, Erie, and Ontario located below. The lakes and their watersheds drain to the Atlantic Ocean through the St. Lawrence River, itself a great waterway (there is also some flow to the Mississippi River, through a canal in Chicago). The aggregate surface area of the Great Lakes is 245-thousand km², and the watershed is 539-thousand km² (Table 20.2). About 35 million people live in the watershed, of which 10-million are in Canada.

All of the Great Lakes except Lake Michigan form part of the border between Canada and the United States. Consequently, issues concerning resources and water quality are binational: waters in Canadian jurisdiction are affected by actions in the United States and vice versa. Especially important issues are the dumping of sewage and industrial waste, the conversion of forest and wetlands into agricultural and residential land-uses, and commercial and sport fishing. Recognizing this context, the governments of Canada and the United States have entered into a number of co-operative agreements regarding the management of resources, emissions of pollutants, and research and monitoring of their shared Great Lakes ecosystem. Much of the integrated binational activity is coordinated by the International Joint Commission, a body with equal representation from both countries. The Great Lakes Water Quality Agreement, implemented in 1972 and modified in 1978, 1983, 1989, and 2012, commits Canada and the United States to maintain and restore the chemical, physical, and ecological integrity of the Great Lakes and their watershed.

Lake Erie has a relatively small volume and is located in a watershed with fertile soil. Consequently, Lake Erie has always been the most productive of the Great Lakes. However, its productivity has been greatly increased as a result of nutrient inputs associated with urban sewage and agricultural drainage. This has created eutrophic conditions in shallower regions of the lake. The western basin of Lake Erie is particularly vulnerable to eutrophication because it is relatively shallow and warm and receives large inputs of sewage and agricultural runoff. In addition to nutrient loading, Lake Erie has been affected by other important stressors. These include contamination by potentially toxic chemicals, large commercial and recreational fisheries, conversion of most of the natural ecosystems in its watershed into agricultural and urban land-uses, and introductions of alien plants and animals. This complex of stressors has degraded the water quality and ecosystem of Lake Erie. The damage was particularly acute during the late 1960s and early 1970s, when pollution was relatively uncontrolled. Although some of the earlier problems have been alleviated, Lake Erie is still in a degraded condition. The following sections examine the most important ecological changes in Lake Erie, as a case study of the effects of eutrophication occurring in combination with other stressors (Freedman, 1995).

Changes in the Fishery

Lake Erie has long supported a large fishery, which typically exceeds the combined landings of all the other Great Lakes. Remarkably, the total catch by the commercial fishery has been quite stable over the years. This has occurred despite enormous changes in the species of fish present, fishing technology, intensity of eutrophication, pollution by toxic chemicals, habitat damage caused by damming rivers required for spawning, and sedimentation of shallow habitat by soil eroded from deforested parts of the watershed.

Although the catch of the fishery on Lake Erie has not declined, the nature of the fish community has changed greatly. These changes illustrate a severe degradation of the fishery resource and of the natural ecosystem. When the commercial fishery on Lake Erie began in the nineteenth century, the prime targets were the largest, most valuable species – these were whitefish (Coregonus clupeaformis), lake trout (Salvelinus namaycush), and herring (Leucichthys artedi). This is a common pattern whenever a previously unexploited fishery or forest resource is initially harvested – take the best and leave the rest (see Chapter 14).

Unfortunately, populations of the most desirable species were rapidly depleted. This happened because the fishing pressure was excessive and could not be sustained. Also, severe habitat degradation occurred in the lake, caused mainly by erosion and siltation associated with extensive deforestation of its watershed. As the most desirable species disappeared, the fishing industry switched to “second-choice” species, such as blue pike (Stizostedion vitreum glaucum), walleye (S. v. vitreum), sauger (S. canadense), and yellow perch (Perca flavescens). Because of over-exploitation and habitat degradation, the species of Stizostedion became extirpated or rare by the early 1970s. The fishery was then dominated by smaller, low-value species such as yellow perch, and by alien fish such as rainbow smelt (Osmerus mordax), freshwater drum (Aplodinotus grunniens), and carp (Cyprinus carpio). Therefore, although the total yield of fish caught has remained fairly large and consistent over time, the quality of the economic resource and the integrity of the fish community have been badly degraded.

Recent Changes in Environmental Quality: For a number of reasons, ecological conditions have improved markedly in Lake Erie since the late 1970s. This has largely been achieved by the construction of sewage-treatment facilities in lakeside cities and towns in Canada and the United States (as well as upstream, especially on the Detroit River and Lake St. Clair). Many of these facilities include technology to reduce phosphorus inputs. The annual loading of phosphorus to Lake Erie has been reduced from about 28-thousand tonnes in 1968, to 20-thousand t in 1974, and 6-17-thousand t during 1981-2008 (average 9-thousand t; Richards, 2012; Figure 20.1). Whereas in the 1960s most of the phosphorus loading was from point-sources (primarily discharges of sewage from towns and cities), it is now mostly non-point discharges associated with agricultural runoff.

The reduction of phosphorus input has alleviated eutrophication in Lake Erie (Makarewicz and Bertram, 1991). The average biomass of phytoplankton decreased from 3.4 g/m³ in 1970 to 1.2 g/m³ during 1983-1985. The largest biomass still occurs in the western basin, where it averaged 1.9 g/m³ during 1983-1987, compared with 1.0 g/m³ in the central basin and 0.6 g/m³ in the deepest, eastern basin. Blooms of nuisance algae have also decreased in intensity. For example, the blue-green alga Aphanizomenon flos-aquae had a standing crop as high as 2.0 g/m³ in 1970, but only 0.22 g/m³ during 1983-1985. Similarly, diatoms that indicate eutrophic conditions have decreased in abundance, by 85% in the case of Stephanodiscus binderanus in the western basin, and by 94% for Fragilaria capucina. At the same time, diatoms indicative of mesotrophic or oligotrophic conditions have become more abundant, notably Asterionella formosa and Rhizosolenia eriensis. The open water of the previously eutrophic western basin is now considered to be in a mesotrophic condition, while the eastern basin is now oligotrophic.

The animal communities have also changed since the 1970s. Species of zooplankton that indicate oligotrophic conditions have become more abundant, while indicators of eutrophication are fewer and mostly restricted to the western basin. Since 1972, the populations of relatively large fishes have increased greatly, particularly walleye and introduced Pacific salmon (Oncorhynchus spp.). These are fish-eating species, and their predation has decreased the abundance of smaller, zooplankton-eating fish such as smelt, alewife (Alosa pseudoharengus), and shiners (Notropis spp.). The decrease of planktivorous fishes has allowed secondary increases of larger-bodied zooplankton, such as the waterflea Daphnia pulicaria.

The zebra mussel (Dreissena polymorpha) is another cause of important ecological change. This bivalve mollusc, a native of Eurasia, was accidentally introduced to the Great Lakes by the discharge of ballast water from transoceanic ships. The mussel can rapidly attain an extremely dense population (up to 50-thousand/m²) on hard underwater surfaces such as rock, metal, and concrete. It is a filter-feeder, and its huge populations have an enormous capability for removing algal cells from water. Consequently, they may be responsible for some of the recent clarification of Lake Erie and eutrophic parts of other Great Lakes (Figure 20.1). In addition, their dense populations have benefited some species of ducks that winter on the lakes and feed on benthic molluscs and other invertebrates. However, the invasion of the Great Lakes by zebra mussels has also caused serious damage, including a reduction in filter-feeding zooplankton (with secondary effects on planktivorous fish), and the displacement of native molluscs that cannot compete with the dense shoals of this non-native mussel. Industries and water utilities have also suffered damage from the clogging of their water-intake pipes.

Hydroelectric Developments

Hydroelectricity is produced by using the kinetic energy of flowing water to turn a turbine, which connects to a generator that produces electricity, which is distributed to consumers through a complex network of transmission lines (see Chapter 13). The global use of hydroelectricity was 3,782 terawatt-hours of electricity (in 2013; this is equivalent to 856 x 10⁶ tonnes of oil equivalent; BP, 2015). About 56% of the global use of hydroelectricity occurs in developed countries, and 23% in North America. Canada has invested relatively heavily in the development of hydroelectric resources and accounts for 10% of the global use of this energy source (for comparison, Canada constitutes 0.5% of the global population). Hydroelectricity accounts for about 6.7% of all commercial energy use, but 75% of the use of renewable sources (others include wind, solar, geothermal, and biomass).

There are three basic ways to harness flows of surface water to generate hydroelectricity:

1. A large impoundment can accumulate riverflow and flood an extensive area of land. This stores water from the spring high-flow period so that electricity can be generated according to demand throughout the year. This kind of facility has a large effect on the seasonal variation of below-dam riverflow, because it greatly reduces the spring peak flow while increasing the flow during the summer, fall, and winter (Figure 20.2). Large impoundments are the most common type of major hydroelectric facility in Canada.
   Figure 20.2. Effects of Reservoir Development on Seasonal Flow of the Peace River. The Bennett Dam and its associated reservoir were built in 1968, so the upper curves (for 1915–1967) represent pre-impoundment conditions. The flows were measured at the town of Peace River, Alberta. The most notable changes are (1) an overall reduction in the annual variation of riverflow; (2) a large reduction in peak flow during mid-April to mid-July; and (3) an increase during the low-flow period of September to March. Source: Modified from Rosenberg et al. (1997).
   

2. A run-of-the-river development directly harnesses the flow of a river to drive turbines, without creating a large impoundment for storage. This kind of facility utilizes riverflow according to its seasonal availability. A run-of-the-river development has little or no capacity to store part of the spring peak flow, or to coordinate the timing of electricity generation with peaks of consumer demand. However, this kind of hydro development causes much less environmental damage than one involving a large reservoir.

3. A combined system incorporates elements of both run-of-the-river and the development of a large impoundment. These so-called peaking systems will store water during part of the day and release it during the time of highest demand for electricity, which is generally between 08:00 and 22:00 (Figure 20.4). One variation is the pumped-storage system, in which electricity generated during low-demand times of the day is used to pump water into an elevated reservoir; the stored water is later used to generate electricity at the peak-demand time. If the peak-flow system causes large flow variations downriver, important ecological damage may be caused. Other combined systems have run-of-the-river generators installed on rivers draining into a central reservoir or installed downriver of a reservoir.

Environmental Effects

Important environmental benefits are associated with the use of hydroelectricity:

• Because the energy source (flowing water) is renewed through the hydrologic cycle, hydroelectricity is a renewable source of power.
• The emissions of greenhouse gases are much smaller than those associated with the use of fossil fuels to generate electricity.
• Unlike fossil fuels, there are no direct emissions of SO₂ or NO_x, which are important causes of acid rain.
• In some regions, impoundments help control downstream flooding, which might otherwise cause economic damage and risks to people living in flood plains (described later).
• Substantial recreational or commercial fisheries may develop in reservoirs.

However, to a degree these advantages are rather simplistic, particularly because large quantities of fossil fuels, metals, and other non-renewable resources are used to construct hydroelectric facilities. Moreover, the dams and impoundments cause some important environmental damages (Dorcey et al., 1997; Rosenberg et al., 1997; International Rivers Network, 2005).

• Flooding of Natural Habitat: Large reservoirs flood extensive areas of terrestrial and wetland habitat. This causes ecological damage, including the displacement of plants and animals that had utilized the original habitats. In some cases, uncommon or rare species may be affected, particularly if unusual habitats such as waterfall spray zones or special wetlands are destroyed. Of course, even while flooding destroys terrestrial and wetland areas, it also develops new aquatic habitats, which provide opportunities for certain fish, waterfowl, and other aquatic species. The productivity of algae, zooplankton, and fish is usually relatively high for several years after the creation of a new reservoir because of nutrients leached from flooded soil.
• Methylmercury: High concentrations of methylmercury commonly occur in fish in reservoirs. This happens because inorganic mercury that is naturally present in soil becomes methylated by bacteria under the anoxic conditions that develop in sediment after flooding. As we learned in Chapter 18, methylmercury is readily bioaccumulated by organisms and then magnifies up the food web to occur in particularly high concentrations in top predators. It also tends to occur in higher concentrations in older individuals within a fish population. Mercury concentrations in the flesh of predatory fish in reservoirs are often higher than 1.0 ppm and can reach 3 ppm, which significantly exceeds the 0.5 ppm limit for fish intended for human consumption. Fish with high concentrations of methylmercury are also a toxic hazard to natural predators such as osprey (Pandion haliaetus), bald eagle (Haliaeetus leucocephalus), and river otter (Lutra canadensis). In the boreal region of Canada, the concentration of mercury in fish increases after the initial development of a reservoir and persists for 10-30 years or longer (Figure 20.5). In general, this phenomenon is most intense in large, new reservoirs. It is less of a problem in reservoirs that are created by raising the water level of a steep-sided lake or river valley with shorter-term storage, or in older impoundments.

  

  • Altered Flows and Obstructions: Many downstream effects result from the construction of dams, other flow-control structures, and diversions of rivers to increase the flow into reservoirs. Changes in the timing and amounts of flow affect the sedimentation regime: decreased flow results in silt deposition and the infilling of gravel beds used by spawning fish, while scouring by increased flow causes other kinds of habitat damage. Large changes in the flow also affect the productivity of algae and macrophytes, in part by affecting the flux of nutrients. These effects can be particularly acute in riverine marshes, deltas, and estuarine habitat. The resulting changes in productivity and other habitat characteristics may secondarily affect aquatic invertebrates, fish, migratory birds, and aquatic mammals. In one rare case in 1984, a torrential flow on the Caniapiscau River in northern Quebec, partly caused by the release of water from an overly full reservoir (due to a week of unusually heavy rain), drowned about 10-thousand migrating caribou (Rangifer tarandus).

    

• Effects of Obstructions: A high dam can be an insurmountable obstruction to the up-river passage of migratory fish. However, this blockage can sometimes be partially mitigated by installing a fish ladder, by catching migrating fish and transporting them above the dam, or by releasing young fish raised at a hatchery. In addition, juvenile fish migrating to the ocean may be killed or injured during passage through the turbines of a hydroelectric facility. Some of this damage can be avoided by installing screens or deflectors, while also providing an alternative passageway for the migrating fish.

  

• Emissions of Greenhouse Gases: The development of a large reservoir results in conditions suitable for the emission of large amounts of carbon dioxide and methane to the atmosphere (see Chapter 17). The production of methane, which is about 28 times more potent as a greenhouse gas than carbon dioxide, may occur if flooding results in anaerobic conditions through the decomposition of large amounts of biomass – the dead trees and litter of drowned forests and the organic matter of inundated wetlands, especially peaty bogs and fens. Such oxygen-poor conditions favour the production of methane during decomposition, which out-gasses into the atmosphere. The emission of greenhouse gases is greatest during the first several decades after flooding and then slows to a rate similar to that of natural lakes. Experimental flooding of a wetland at the Experimental Lakes Area resulted in a twenty-fold increase in the rate of methane emission to the atmosphere. Under conditions that are particularly favourable for methane generation, the rate of emission of greenhouse gases (standardized as greenhouse warming potential because CH₄ has greater radiative activity than CO₂) can exceed that of a coal-fired power plant (although this is more characteristic of tropical reservoirs than cooler ones; Rosenberg et al., 1997).

• Effects on Biodiversity: Some hydroelectric developments have destroyed the habitat of threatened species or unusual ecosystems. The proliferation of dams on rivers has greatly reduced or extirpated many populations of Atlantic salmon (Salmo salar) and Pacific salmon (Oncorhynchus spp.). At least 142 stocks of Pacific salmon in western Canada have been lost partly because of hydroelectric dams (in combination with logging, over-fishing, and other stressors), and many others are threatened. Hydroelectric developments also threaten the breeding habitat of the harlequin duck (Histrionicus histrionicus) in both eastern and western Canada (it is endangered in the eastern part of its range). Construction of the Churchill Falls hydro project in Labrador destroyed the habitat of rare species of ferns, mosses, and liverworts in the misty spray zone of the original natural waterfall. Undoubtedly, numerous undocumented losses of rare species and their habitat occurred during the construction of hydroelectric developments in Canada and elsewhere prior to the early 1970s, when biodiversity surveys became a routine component of environmental impact assessments for these projects.

• Effects on Local People: The lifestyle of local people can be greatly affected by a hydroelectric development. This is particularly true if a large reservoir is developed in a heavily populated area, which is often the case in less-developed countries. For example, in China alone, the Three Gorges Dam displaced about 1.3-million people; the Danjiangkou Dam (completed in 1974) displaced 383-thousand people; the Sanmenxia (1960), 319-thousand; the Xinjiang (1961), 306-thousand; and in India, the Dongpinghu Dam (1958) displaced 278-thousand people (Goodland, 1994). In such crowded countries, suitable land for relocation of these displaced people is often unavailable. People are also displaced by reservoirs in northern Canada, although typically several hundred or fewer. Almost all of the displaced people are Aboriginal, who may have to be relocated from traditionally used areas if their villages become flooded. They are also deprived of opportunities to hunt mammals and birds in part of their traditional foraging area, and they may not be able to eat fish from the reservoir or downstream for several decades after its creation (because of health hazards associated with methylmercury). Of course, local economic opportunities exist for some of these people, such as jobs associated with the construction and maintenance of the hydro facility and its related infrastructure of transmission lines and roads. However, entry into wage employment can be extremely disruptive to traditional subsistence lifestyles, for both individuals and the community. There are also many social and economic disruptions caused by the influx of people from elsewhere in Canada and from the construction of new roads and towns. For these and other reasons, local people often bitterly resist the development of large hydroelectric facilities (and other large industrial projects) in the areas where they live.

Canadian Focus 20.2. Hydro Development in Labrador
Hydroelectric power is an economically attractive and renewable source of energy. In the past, areas with a high potential for hydroelectricity were often developed without much consultation with local people, who also may not have received many economic benefits from the project. Local people generally did, however, bear the brunt of the environmental damage, which typically included extensive flooding, changes in the hydrology of rivers used for transportation or for fishing, and sometimes harmful influences on traditional lifestyles and culture.

One example of insensitive hydro development was the Churchill Falls Project in Labrador, for which construction began in 1967. In 1969, the government of Newfoundland entered into a contract with Hydro-Québec (a Crown Corporation of Québec) to supply power for 65 years, but at a price that (incredibly, in retrospect) took no account of monetary inflation or the future value of hydroelectricity. In 1971, flooding of the Smallwood Reservoir began, and it eventually covered an immense 6,700 km². In 1974, a generating station with a capacity of 5429 MW was completed and electricity began to flow to markets in Québec and the northeastern United States.

It is astonishing by the standards of today, but the Churchill Falls Project was developed with only a few studies of its potential environmental impacts. For instance, there was little understanding of how the local and regional ecology might be affected by the enormous changes in the hydrology of the Churchill River, one of the great watercourses of eastern Canada. Inevitably, the huge reservoir caused extensive damage through flooding, other habitat changes, and the mobilization of bioavailable methylmercury.

It is also remarkable that this immense industrial development occurred without much consultation with the local people, particularly those of the Innu culture, who are the original inhabitants of that region of Labrador. The local Innu had long engaged in subsistence hunting and commercial trapping in the project region, undertaking seasonal movements to traditional areas where they obtained wild meat as food and furs to sell. Most of these people were completely unaware of the proposed hydroelectric development or its likely consequences for their traditional activities. Some of them lost their canoes, cabins, and other possessions to flooding, for which they were not compensated. Many archaeological sites, such as burying grounds, were also lost.

In 1975, planning and construction began for a subsequent hydroelectric development, this time to harness the lower Churchill River by building a dam and power plant at Gull Island, about 200 km downriver of Churchill Falls. Although the project was not completed at the time, the development potential remained attractive in terms of industrial and economic considerations. In 1990, and several times since, new proposals were brought forward to develop the hydroelectric potential of the lower Churchill River, and as of 2015 (when this case study was written), it looked certain that the project would proceed. Although these plans have generally been supported by business interests in nearby Goose Bay, many of the Innu have been opposed to the proposed new hydro project. Their governing organization, known as Innu Nation, has demanded the following features in any development agreement they would consider signing:

• compensation for damage suffered by Innu during the initial development at Churchill Falls
• a full assessment of the environmental and socio-economic impacts of the proposed new development
• royalties and other compensation, such as specified job and contract opportunities, for Innu persons and companies during the new development
• the settlement of a comprehensive land claim, in view of the assertion of the Innu Nation of Aboriginal title for almost all of that region of Labrador

To reinforce their position, the Innu have at various times held public protests in the project area and on Newfoundland. They have also petitioned their case to environmental organizations, power companies, and state governments in the northeastern United States, where much of the power would likely be sold. A particularly high-profile demonstration was held to confront a press conference staged by the governments of Québec and Newfoundland in March of 1998, near the village of Churchill Falls, where their premiers jointly announced a new plan for a hydro development on the Lower Churchill River. This protest by the Innu garnered front-page attention throughout Canada and internationally. The project envisioned at that time would have cost about $10 billion. It would have constructed a new reservoir, dam, and generating facility of 2,264 MW at Gull Island, while also increasing the capacity of the existing Upper Churchill Falls station by 1000 MW.

Since then, the government of Newfoundland & Labrador has enetered into an agreement with Nova Scotia to develop a hydro facility on the Lower Churchill River and a transmission corridor through to New England. However, at the time of writing (2015), full consensus had not yet been obtained from the Innu people, and a land settlement had not been achieved. However, many detailed or environmental and socio-economic impact assessments had been completed. Although this hydroelectric development in Labrador remains contentious, its construction is likely to begin in the summer of 2015.

A Case Study: La Grande Complex

The La Grande Complex in Québec was developed between 1973 and 1996 (Messier, 1998). The development is centred on the La Grande River, but the river’s natural flow has been augmented by diversions of the Caniapiscau River (48% of its flow) and the linked Eastmain and Opinaca Rivers (90% of their flow). The first phase of the development occurred between 1973 and 1985 and resulted in five reservoirs with a total impoundment area of 11,335 km² (including 10,400 km² of newly flooded land). Three powerhouses were built, with a total generating capacity of 10,282 MW. The second phase of the development, between 1987 and 1996, added five powerhouses (capacity 4,962 MW) and three new reservoirs (1,618 km², including 1,134 km² of flooded land). During the planning, construction, and operating phases of the development, Hydro-Québec undertook extensive studies of hydrology, climate, ecology, socio-economics, and other issues related to potential environmental impacts.

For several years after the new impoundments were flooded, there was a relatively high productivity of phytoplankton and zooplankton. This occurred because of the high concentrations of nutrients, especially phosphorus, that were leached from flooded soil. The resulting abundance of invertebrates and small fish allowed a relatively high productivity of lake whitefish (Coregonus clupeaformis) and northern pike (Esox lucius) for several years. However, mercury concentrations up to 3 ppm developed in these and other predatory fishes. The maximum concentrations in whitefish occurred five years after impoundment and subsequently declined, while those in pike peaked after 10 years. The mercury concentrations are expected to return to background levels after 10-25 years for fish that feed on invertebrates, and after 20-30 years for fish that feed on other fish.

Erosion associated with high-water discharges threatened the village of Fort George on the estuary of the La Grande River, forcing the relocation of its inhabitants to a new settlement upstream at Chisasibi. The relocation resulted in considerable social and lifestyle disruptions for the people, almost all of whom were Cree. The immense impoundments also affected the patterns of land-use by local hunters and trappers, some of whom no longer had access to their traditional areas. Greatly increased winter flows associated with hydroelectricity production have also created unstable ice conditions on the river. Consequently, some former winter travel routes are no longer safe to use.

Local people and their communities were also greatly affected by the diverse economic effects of the land-claim settlement and seven project-related agreements. (A total of $555 million was allocated for compensation and remedial work on Cree, Inuit, and Naskapi lands associated with the development of the La Grande Complex.) They were also greatly affected by the entry into wage employment, the construction of a network of roads and other infrastructure, the influx of many non-local people working on the hydro facilities, and by other rapid socio-economic changes. Some of the changes have been viewed favourably but others have not, particularly if they are considered to have degraded the traditional elements of Cree, Inuit, or Naskapi culture.

Impoundments for Flood Control and Irrigation

Some areas are highly vulnerable to flooding in the springtime, especially if the yield of water from the watershed due to rapid snowmelt or a severe rain event exceeds the capacity of a river channel, which may cause massive spillover onto normally terrestrial habitat. An area that is vulnerable to this kind of hydrological influence is known as a flood plain, and they are common in many parts of Canada (see Canadian Focus 20.3). Such flooding may occur regularly or it may happen only in years with unusually high water yield from the watershed. To prevent or reduce the damage caused by flooding, control structures such as dams, reservoirs, and channelled spillways may be constructed.

Some Canadian proposals to develop impoundments for flood control and irrigation have been extremely controversial. Two of the most notable have been the Rafferty-Alameda project in southeastern Saskatchewan and the Oldman Dam in southwestern Alberta.

The Rafferty-Alameda project consists of the Rafferty Dam and Reservoir on the main branch of the Souris River, the Alameda Dam and Reservoir on Moose Mountain Creek, and associated channelizations and diversions. This project was undertaken to provide the following benefits: water for the irrigation of about 4,800 ha of land, regional flood control, storage of municipal water, cooling water for a thermal power plant, and reservoir-based recreation. However, opponents of the project objected to some of its environmental impacts, which included the displacement of 75 farm families by flooding of the reservoirs, the destruction of habitat of rare prairie plants and animals as well as fish, a degradation of downstream water quality and quantity, and damage to cultural and historical sites. Intense controversy was engendered by this proposal, and a legal challenge forced it to undergo a full-scale environmental impact assessment (EIA). This was done, even though construction activities were already at an advanced stage and had to be temporarily halted. The Rafferty Dam was completed in 1992 and the Alameda, in 1994.

It should be noted that the Rafferty-Alameda Project was the subject of a landmark decision in Canadian law, known as the Rafferty Decision. The project was initially issued a permit by the provincial government, but its opponents initiated legal action based on the need for a full EIA under the requirements of federal law. In 1989, the Federal Court of Canada ruled that the Department of the Environment (a federal agency) must conduct a comprehensive EIA of the proposed project under the Environmental Assessment and Review Process (EARP). That legal decision gave the federal EARP guidelines the force of law, making it mandatory to do such assessments for any development proposals involving the federal government. This includes any federally funded proposal, even if the development is to be undertaken by another (non-federal) organization, or if there is a risk of an environmental effect on an area of federal responsibility. Much of the political controversy in this case was associated with disagreements between the provincial and federal governments over jurisdictional issues.

The Oldman project consists of a dam, a reservoir, and associated structures on the Oldman River. The project was undertaken to provide the following benefits: water for irrigation of 68,850 ha of land, regional flood control, storage of municipal water for Lethbridge and Fort McLeod, enhancement of downstream fish populations through flow regulation, and reservoir-based recreation. However, opponents of the project, including members of the local Peigan Aboriginal Nation, objected to some of its likely environmental impacts, which are similar to those described for the Rafferty-Alameda case. Opponents of the Oldman Dam were successful in a legal action to force the proponents to undertake a full EIA (this was the first application of the Rafferty decision of 1989; it was eventually resolved by the Supreme Court of Canada). The EIA was conducted, even though the development was already about 40% completed. Construction of the dam was completed in 1993. However, much controversy remained because some concerns of the environmental review panel were not fully addressed before the permit was granted for completion.

Canadian Focus 20.3. Disastrous Floods
Severe flooding is a natural catastrophe that can disrupt the lives of many people, even killing them, and causing expensive damage to homes, agriculture, and industry. Flood-control structures, such as dams, reservoirs, and levees, can help to prevent overflowing into vulnerable low-lying areas, but they are not fail-safe. Three famous episodes of recent catastrophic flooding in Canada occurred in the Saguenay region of Quebec, in southeastern Manitoba, and in southern Alberta.

The Saguenay disaster, in 1996, was caused by an intense July rainstorm that delivered 15.5 cm of precipitation during a 50-hour period. The resulting flooding caused the deaths of 10 people and forced about 16-thousand others from their homes. About 1,350 houses were destroyed or badly damaged, and one-thousand families had to be permanently relocated. There was also a devastating impact on commercial, industrial, and tourism operations. The total economic damage was more than $800 million. Although some of the rivers feeding into the Saguenay region had flood-control dams and reservoirs in place, these were not adequate to deal with the enormous volume of water generated by the extreme rain event.

A much-more extensive event affected the flood plain of the Red River in 1997, as a result of a high volume of spring runoff from an unusually deep snowpack that had accumulated from several storms. The flooding affected more than 2-thousand km² of flat low-lying terrain, and it caused 24-thousand people to be evacuated from their homes in southeastern Manitoba and more than 50-thousand in North Dakota and Minnesota, including almost all of the population of the city of Grand Forks. The total cost of the damage from this flood exceeded $1 billion. The destruction in Manitoba would have been much worse, but an extensive network of temporary dikes was hurriedly built by local residents, volunteers, and the Canadian army. Along with an existing system of permanent dikes and diversion channels, these protected most of the city of Winnipeg, which otherwise would have been devastated by flooding (although these same dikes did exacerbate some of the flooding that occurred south of Winnipeg). Crucial to saving the city was the Red River Floodway, built in 1968 to give protection from just this kind of severe flooding. Until the great flood of 1997, many people somewhat derisively referred to this floodway as “Duff’s Ditch,” after Duff Roblin, the premier who had authorized construction of the project against intense political opposition because of its high capital costs.

The third case occurred in the early summer of 2013, when heavy rains triggered the worst flooding ever recorded in southern Alberta. The rainfall reached 32.5 cm over two days in some places on the eastern slope of the Rocky Mountains, and the massive downward flow from large watershed then inundated low-lying floodplains of the Bow, Elbow, Highwood, Red Deer, Sheep, Little Bow, and South Saskatchewan rivers and their various tributaries. More than 100-thousand people were displaced from their homes, four people died of drowning, and the economic damage was estimated to exceed $5 billion.

Image 20.6. A view of flooding in Calgary, looking towards the East Village neighbourhood during the Great Alberta Flood of 2013. Source: Ryan L.C. Quan, Wikimedia Commons, http://en.Wikipedia.org/wiki/2013_Alberta_floods#mediaviewer/File:East_Village_Calgary_Flood_2013.jpg