Info

Aquaponics 4 You

Aquaponics Projects For Beginners

Get Instant Access

FIGURE 13.7 Flows of the Colorado River below all major dams and diversions, 1904-2004. (Reproduced from Postel 2005.)

13.2.1.3 Land use

Land-use change is an integrator of many human activities that have a negative impact on stream ecosystems. Flow variability and sediment delivery to streams are strongly influenced by the impervious surfaces and storm drains of urban areas, the channelized streams and field tiles of agricultural areas, and logging practices and road building in areas of forest harvest. Habitat degradation is likely to be manifested in changes to flows, benthic habitat conditions, and riffle-pool integrity. In regions that naturally have a vegetated riparian, the presence of vegetation stabilizes banks, moderates water temperature, filters nutrients and sediments, and strongly influences energy pathways by influencing the availability of light and inputs of particulate organic matter (Gregory et al 1991). Typically the loss of riparian vegetation is accompanied by bank erosion, silt deposition, warmer water, and altered food webs; and as human presence encroaches to the stream edge contaminant loading often increases (Lowrance et al. 1997, Hickey and Doran 2004). Although the pathways are less obvious, land use patterns and human presence throughout the catchment often serve as good indicators of a stream's condition, acting as a general index of human disturbance (Table 13.3).

Most settled regions of the world have seen extensive transformations of their landscapes. In New Zealand, over 80% of the land was forest before agricultural expansion; today, pasture for sheep is the dominant land use in the middle and lower catchment areas of most of New Zealand's streams and rivers (Quinn 2000). Agriculture is the dominant land use in many developed watersheds in the United States, comprising >40% of the land area of the Lower Mississippi, Upper Mississippi, Southern Plains, Ohio, Missouri, and Colorado River basins (Allan 2004). Urban land use typically makes up a lower percentage of total catchment area, and for large basins urban land is usually <5% of catchment area. Because urban stressors have a disproportionate influence on aquatic ecosystems (Paul and Meyer 2001) the influence of urbanization can be important even at low values. When small catchments of low-order streams are the focus,

TABLE 13 3 Principal mechanisms by which land-use activities influence stream ecosystems. (From Allan 2004.)

Environmental factor

Effects

Sedimentation

Nutrient enrichment

Contaminant pollution

Hydrologie alteration

Riparian clearing/ canopy opening

Loss of large wood

Increases turbidity, scouring, and abrasion; impairs substrate suitability for periphyton and biofilm production; decreases primary production and food quality causing bottom-up effects through food webs; in-filling of interstitial habitat harms crevice-occupying invertebrates and gravel-spawning fishes; coats gills and respiratory surfaces; reduces stream depth heterogeneity leading to decrease in pool species

Increases autotrophic biomass and production, resulting in changes to assemblage composition, including proliferation of filamentous algae, particularly if light also increases; accelerates litter breakdown rates and may cause decrease in dissolved oxygen and shift from sensitive species to more tolerant, often nonnative species

Increases heavy metals, synthetics, and toxic organics in suspension, associated with sediments, and in tissues; increases deformities; increases mortality rates and impacts to abundance, drift, and emergence in invertebrates; depresses growth, reproduction, condition, and survival among fishes; disrupts endocrine system; physical avoidance

Alters runoff-evapotranspiration balance, causing increases in flood magnitude and frequency, and often lowers base flow; contributes to altered channel dynamics, including increased erosion from channel and surroundings and less-frequent overbank flooding; runoff more efficiently transports nutrients, sediments, and contaminants, thus further degrading instream habitat. Strong effects from impervious surfaces and stormwater conveyance in urban catchments and from drainage systems and soil compaction in agricultural catchments

Reduces shading, causing increases in stream temperatures, light penetration, and plant growth; decreases bank stability, inputs of litter and wood, and removal of nutrients and contaminants; reduces sediment trapping and increases bank and channel erosion; alters quantity and character of dissolved organic carbon reaching streams; lowers retention of benthic organic matter owing to loss of direct input and retention structures; alters trophic structure

Reduces substrate for feeding, attachment, and cover; causes loss of sediment and organic material storage; reduces energy dissipation; alters flow hydraulics and therefore distribution of habitats; reduces bank stability; influences invertebrate and fish diversity and community function land use can vary from nearly 0-100% coverage of urban, agricultural, or forested land.

When streams that drain catchments under different land use are compared, the influence of landscape setting is apparent. Quinn et al. (1997) compared water quality, habitat, and biota in 100 m reaches of New Zealand streams draining pasture, native (podocarp-broadleaf) forest, and exotic pine forest that previously was pasture. Pasture streams received much more light, were warmer, had higher nutrient levels, and much higher algal biomass. Both pine and pasture streams had more fine sediment stored in the streambed than native streams; pine streams had the most wood, and pasture streams the least. Community composition differed most between pasture and native forest, with more midges and snails in the former, and more mayflies, stoneflies, and caddis-flies in the latter.

Many studies have shown that habitat quality and biological diversity (often assessed using metrics of biological integrity, see Section 13.3.1) correlate with various measures of land use either along the river corridor or throughout the stream's catchment (Allan 2004). Based on a comparison of 30 streams in the Etowah River basin, Georgia, Roy et al. (2003) found that number of taxa and other biotic indices were negatively related to urban land cover and positively related to forestland cover. Urban land cover correlated negatively with streambed sediment size and positively with nutrient concentrations and turbidity, suggesting that environmental conditions affecting the biota reflected land use. A similar study of 134 stream sites throughout Wisconsin found that habitat quality and biotic integrity were negatively correlated with the amount of agricultural land and positively correlated with the amount of forested land (Wang et al. 1997).

There is ample evidence that poorly regulated forest harvest has resulted in substantial degradation in habitat and fish populations (Bisson et al. 1992). Changes in streamflow and increased sediment production are among the most serious consequences of logging activities because they have long-term effects on channel and habitat features. Major sources of sediments include landslides from deforested slopes, surface scour from logging roads, and erosion of sediments stored on stream banks or within the streambed itself due to greater flooding (Scrivener and Brownlee 1989, Bruijnzeel 2004). As more of the basin is logged, pools fill with sediments, pool-forming large wood is lost, and the frequency and size of pools decline (Bisson et al. 1992). In general one observes a reduction in species diversity, attributed to habitat simplification, and an increase in standing crop biomass, attributed to greater light penetration and auto-trophic production. Surveys of streams in logged versus unlogged watersheds in the Pacific Northwest found that salmonid species diversity was lower in logged areas regardless of underlying geology (Figure 13 8). Aquatic invertebrates also are adversely affected by clear-cut timber harvest. Corn and Bury (1989) compared amphibian abundances in headwater streams in uncut forests to streams in second growth forests that had been logged between 14 and 40 years previously. Species richness was lower and the percentage of fine sediments was greater in previously logged streams, demonstrating the long-term consequences of timber harvest.

How effectively a riparian buffer strip can lessen the adverse impacts of land use on aquatic ecosystems, and what width is required, continue to be debated. The recommended width of buffer strips varies from <10 to 100 m, and will likely depend on protection goals and context (Lee et al. 2004). Attempts to distinguish between the importance of riparian versus catchment land use to stream condition and the effectiveness of riparian buffers of various widths have produced mixed results, in part because land cover variables are often correlated (Roth et al. 1996). By selecting sites with uncor-related forest cover values and developing a model relating buffer width to stream variables, Jones et al. (2006) demonstrated that the reduction in buffer width from 30 to 15 m results in

FIGURE 13.8 Diversity of salmonid fishes in logged (shaded bars) and unlogged (unshaded bars) Oregon coastal streams with different parent rock types. Diversity is expressed as H' (—Spj log pi, where pi is the frequency of the ith species). (Reproduced from Bisson et al. 1992.)

elevated stream temperatures and reduced trout populations in Georgia streams. Even when a riparian buffer exists, its effectiveness can be counteracted by stormwater drainage pipes, subsurface drainage tiles in farm fields, and gullies in eroded landscapes, which rapidly convey water, sediments, and pollutants into stream channels. In urban streams in Melbourne, Australia, invertebrate populations were more highly impacted when storm drains conveyed runoff directly into streams, compared with urban areas that had considerable impervious surface but lacked direct connection (Walsh et al. 2005a, b).

13.2.2 Nonindigenous species

Nonindigenous, exotic, and alien are terms applied to species that colonize a region where they do not naturally occur. Nonindigenous species that become abundant in new environments are referred to as invasive, implying harm to the recipient community. Alien species are of concern for several reasons. Once established in a new environment, they usually become permanent residents and often are capable of reproducing and dispersing far beyond the point of origin. In contrast to chemical pollutants that can be eliminated at their source, or habitats that might potentially be restored, species introductions usually are impossible to reverse. Natural enemies may be lacking, and the impacts of introduced species in new habitats are highly unpredictable because of differences in the nature of species interactions under novel ecological conditions.

Introductions are well documented for a wide variety of freshwater plants, invertebrates, and fishes. A compilation of international introductions of inland aquatic species as of 1988 recorded a total of 1,354 first introductions of 237 animal species, primarily fishes, into 140 countries (Welcomme 1988). The majority of occurrences are of limited extent. Most species have been introduced to ten or fewer countries, and 40% have been recorded from only one country. At the other extreme, nine species have been introduced into more than 30 countries, including three popular sports fish (rainbow and brook trout, largemouth bass); Gambusia affinis, used in mosquito control; and two tilapias and three species of carp popular for aquaculture and weed control. Nonindigenous species often are additions to the receiving community, which may retain all or most of its complement of native species. However, whenever the majority of the biomass consists of introduced species, changes in food web structure and ecosystem function are highly probable. At least 72 nonindigenous species have been successfully introduced in the Colorado River system, which originally held 49 native species (Blinn and Poff 2005), and within the Grand Canyon National Park, over 85% of individual fish are nonnative (Minckley 1991).

Invasive plants including aquatic macrophytes and riparian species also take their toll of aquatic ecosystems by reducing recruitment of native plants and modifying ecosystem processes and disturbance regimes (Gordon 1999). The water hyacinth Eichornia crassipes, originally from South America and introduced into the United States in the late 1800s as an ornamental, is considered one of the world's worst aquatic weeds in subtropical freshwaters. It occupies large areas of the Florida everglades, where its extensive floating mats impede flow, eliminate native plants, and reduce oxygen levels (Schmitz et al. 1993). Salt cedar (Tamarix spp), native to southern Europe and Eurasia, has colonized extensive riparian areas of the US southwest, where it displaces native riparian species and increases water loss due to its high transpiration rates (Cleverly et al. 2006).

13.2.2.1 Causes of species invasions

Freshwater species are purposefully transported and released in new environments to enhance sports fishing, for aquaculture, and as agents of biological control. Indeed, stocking of nonnative fish remains a cornerstone of management efforts to provide the most desirable species for sport or commercial fisheries. Of the 110 documented invasions of fish species into California, 45% were introduced for food or sports fishing and 15% as forage for sports fishes (Moyle and Marchetti 2006). Although relatively few species are extensively used in aquaculture (Welcomme 1984), these are among the most widely transferred fishes. Introductions associated with fish culture expanded considerably in the 1960s, as international development agencies promoted aquaculture to provide protein for rapidly expanding human populations and to benefit local economies. Fishes also have been introduced as biological control agents to combat disease vectors and noxious aquatic weeds. Mosquito control has been a frequent objective, using species such as mosquitofish (Gambusia affinis) and guppies (Poecilia reticulata). Control of aquatic weeds, many of which themselves are exotic, is another frequent objective (Shire-man 1984), with tilapia and carp the most commonly promoted species.

Unintentional introductions occur in a number of ways, including escapees from fish farms, the release of aquarium pets, unnoticed species that "hitch-hike" with a planned introduction, and those carried in ballast water or dispersed through canal systems. Of the 1,205 introduction records for aquaculture purposes listed in FishBase (an encyclopedic database of finfish biology), about half are reported to have become established in the wild (Casal 2006). Over 1,200 species of tropical aquarium species are shipped to various parts of the world, and as many as 6,000 species may ultimately be of interest to the pet trade overall (Welcomme 1984). Most tropical fish introductions have occurred since 1960, when techniques greatly improved for the live transport of fish. About 10% of international transfers of exotic fishes have been the result of truly nonpurposeful introductions, such as accidental transfers of small cyprinids included with shipments of juvenile carp species. Recently, ballast water introductions have been a focus of concern, illustrated by invasions into the Laurentian Great Lakes of the ruffe (Gymnoce-phalus cernua), the zebra mussel (Dreissena polymorpha), and the mitten crab (Eriocheir sinensis) (Ricciardi and Maclsaac 2000). Inter-catchment water transfers have contributed to species translocations in southern Africa because almost all of that region's major river systems are connected by tunnels, pipes, and canals (Bruton and van As 1986). At least five fish species have invaded the Orange River from the Great Fish River, which prior to their connection had distinct faunas with high numbers of endemics.

13.2.2.2 Invasion success

Invasion success is expected to be influenced by traits of the invading species and characteristics of the receiving environment, including "environmental resistance'' from the invaded community. The successful establishment of an alien species can be viewed as a succession of probabilistic events, beginning with transport or dispersal and followed by establishment and further spread, until finally becoming fully integrated into the community (Figure 13 9). Lastly, its impact on native species and ecosystem processes determines whether the novel species qualifies for nuisance or invasive status.

Marchetti et al. (2004a) undertook a detailed analysis of invasion success for freshwater fishes of California, which has 68 native species and 110 documented nonnative invasions. Because 43% of these invasions were failures, it was possible to explore correlates of the successes. The best statistical predictors of a species' invasion success included a past history of successful invasion, broad environmental tolerance, ability to thrive in human-altered environments, similarity of source and recipient environments, and large propagule size (at least 100 individuals, multiple releases, or both). Prior success was a particularly good predictor of the ratio of success to failure, which was near 1:1 for species

FIGURE 13.9 The species invasion process as envisioned for fishes in California: at each step, invading individuals encounter factors that restrict some or all from proceeding to the next step. Successful invasion requires surviving each stage of the process. Once a species is established, a combination of biotic and environmental factors determines whether it will spread and eventually be integrated into the existing assemblage of species. (Reproduced from Moyle and Marchetti 2006.)

FIGURE 13.9 The species invasion process as envisioned for fishes in California: at each step, invading individuals encounter factors that restrict some or all from proceeding to the next step. Successful invasion requires surviving each stage of the process. Once a species is established, a combination of biotic and environmental factors determines whether it will spread and eventually be integrated into the existing assemblage of species. (Reproduced from Moyle and Marchetti 2006.)

with no previous record of successful establishment in a new environment, rising to 4:1 for alien species established in ten or more countries. Success rates were high for game and forage fish, with a 2:1 success to failure ratio. In contrast, unintentional releases of aquarium pets, fish carried in ballast water or moving through constructed waterways, and escapees from fish farms had a 28% success rate. High physiological tolerance was found in 66% of all successful invasions in California (Marchetti et al. 2004b), but was no guarantee of success as 47% of failures had similar high tolerances. What is clear, of course, is that species invasions fail when environmental conditions exceed their tolerances, as is the case for many tropical aquarium fishes in temperate waters. At least some invaders appear to be successful due to novel traits, such as feeding on underutilized resources, high aggressiveness, or timing of reproduction. Species with wide native distributions might be expected to be more successful colonists but this was reported to be at best a weak indicator, and other candidate characteristics such as small body size, fast life cycle, and generalist trophic position met with even less success in the California case study. At this time we can say that some traits are known to predispose a species to become invasive, but predictive models have their limits and so it is advisable to err on the side of caution in risk assessments (Moyle and Marchetti 2006).

Environments that are highly similar to those found within the native range of the invading species are expected to offer the greatest opportunity for successful establishment. The rainbow trout is an excellent model to evaluate environmental factors that might influence colonization success because it is widely introduced (Figure 1310) and good records of introductions often are available (Fausch et al. 2001). They have been highly successful in the southern Appalachians, but most introductions to Scandinavia, central Europe, the United Kingdom, and the main island of Japan have failed. In other areas, including the Rocky Mountains, the Andes, Australia, and New Zealand, rainbow trout introductions have met with moderate success. By extensive analysis of flow records, Fausch et al. (2001) make a strong case that the coincidence of the timing of fry emergence with a low probability of flooding characterizes their native environment and strongly predicts success at establishment; in other words, the match between the native and receiving habitats is critical.

Habitat alteration can facilitate species invasions by creating novel environmental conditions to which native species are not well adapted. When free-flowing rivers are transformed into lakelike impoundments by dams and river regulation, native species may decline because of these changes and successful invaders will likely be those best suited to the altered ecosystem. A survey of 128 species of introduced freshwater fishes across 125 major drainages in North America found that large drainage basins with many impoundments and fewer native species contained the greatest number of introduced species (Gido and Brown 1999). Working with an extensive suite of morphological, behavioral, and life history traits for all fish species (28 native, 62 nonnative) of the Lower Colorado basin, Olden et al. (2006) found that rapidly spreading invasive species tended to occupy "vacant" niches, particularly with regard to life history traits, and the displaced species tended not to overlap in life histories with successful invaders. Thus the opportunity to invade previously unoccupied niche space, most likely arising from long-term environmental changes

FIGURE 13.10 Transfer of rainbow trout (Oncorhynchus mykiss) from its original range in western North America (shaded area) to every continent but Antarctica. (Reproduced from Petersen et al. 1987.)

and particularly the establishment of lentic conditions in impoundments, has played a major role in the establishment of invasive species in the Colorado River Basin.

Finally, the three most successful invaders of freshwater worldwide, the common carp (Cyprinus carpio), rainbow trout (Oncor-hynchus mykiss), and a tilapia (Oreochromis mossambicus), provide interesting contrasts in the degree to which invasive species duplicate the ecological role of a native species, or add a novel dimension to the assemblage. Rainbow trout (and brown trout as well) often displace another salmonid or a troutlike form. On the other hand, both carp and tilapia have been successful in regions where it appears that the ecological or life history traits of the invading species differs from all members of the recipient assemblage. Based on a comparison of species traits of carp to abundant native fishes in Australia, Koehn (2004) argued that carp were clearly different in their resource use and behavior, as detritivorous fishes are lacking in most freshwater fish assemblages in Australia. The success of Oreochromis is attributed to its broad physiological tolerance, extended reproductive season, protection of young, mouth-brooding behavior that allows them to reproduce in all habitats, omnivorous feeding, and predation upon young of coexisting fish species (Canonico et al. 2005).

13.2.2.3 Impacts of invasive species

The impact of nonindigenous species on the native biota is highly variable. Evidence to date suggests that many, perhaps the majority, of invading species have little impact, do not become abundant, and are assimilated into the existing assemblage without causing significant change. On the other hand, some invasive species clearly are harmful, and their contribution to species imperilment in freshwater is perhaps second only to habitat loss (Harrison and Stiassny 1999). Some 167 of Mexico's roughly 500 spe cies of freshwater fishes are listed as being at some degree of risk, and 76 of these are attributed at least in part to the influence of invasive species (Contreras-Balderas et al. 2002). A survey of 31 case studies of fish introductions to stream communities in Europe, North America, Australia, and New Zealand found that 77% of the cases documented a subsequent decline in the native species (Ross 1991). Examples included the decline of native species in the southwestern United States following the introduction of mos-quitofish, and declines of the native brook trout following the introductions of brown and rainbow trout. Trout and galaxiid fishes in New Zealand are incompatible; formerly widespread populations of galaxiids are now fragmented into remnant populations restricted to regions above barrier waterfalls inaccessible to trout (Townsend and Crowl 1991). In addition, nonin-digenous species may cause a variety of indirect effects via food web interactions. In New Zealand, introduced trout exert greater top-down control over invertebrates than the native galax-iids, resulting in a reduction in benthic grazing and an increase in algal biomass (Flecker and Townsend 1994).

Declines in native species following fish introductions occur via a number of mechanisms, including species interactions, habitat alterations, introductions of diseases or parasites, trophic alterations, and hybridization (Taylor et al. 1984). Predation appears to be a common cause of the replacement of native species by exotics. Of ten studies reviewed by Ross (1991) where resource use was examined, habitat shifts following fish introductions were observed in half. For example, a variety of native species including the Sacramento sucker, rainbow trout, California roach, and three-spined stickleback shifted patterns of habitat use in the presence of the Sacramento squawfish, a predatory cyprinid introduced into the Eel River of California (Brown and Moyle 1991). Invading species also affect native species by hybridization, which Miller et al. (1989) found to be a factor in 38%

of the recorded extinctions of North American fish species. In most instances some other factor apparently resulted in the initial decline, and hybridization was the final blow. Examples include at least two subspecies of native cutthroat trout that have gone extinct because of interbreeding with stocked rainbow trout, the Snake River sucker, which hybridized with the Utah sucker, and the blue pike, which hybridized with the walleye.

A host of diseases and parasites are associated with alien species (Hoffman and Schubert 1984, Bruton and van As 1986), posing yet another threat to the invaded community. A fungal parasite causing crayfish plague decimated native crayfish throughout Europe following the introduction of resistant crayfish species from North American (Reynolds 1988). The cestode Bothriocephalus acheilognathi, which originated in China and the Far East and has been introduced worldwide with grass carp, has the potential to affect a wide variety of species (Bru-ton and van As 1986). Whirling disease, caused by Myxobolus cerebralis, was introduced to the United States from Europe in the 1950s and now affects wild fish and fish hatcheries in 23 states (Bartholomew and Reno 2002).

The worldwide spread of invasive species coupled with the decline or extirpation of native species results in a fauna that is becoming progressively more similar across regions, which in turn lessens the uniqueness of local faunas. This is the phenomenon known as biological homog-enization, and freshwater fishes are a prime example. Using historic and present distributions of fishes of the 48 contiguous American states, Rahel (2000) compared the fish faunas for all states (1,128 pair-wise combinations), finding a clear increase in faunal similarity. On average, pairs of states have 15 more species in common than historically, and the 89 state pairings that originally had no fish species in common now have an average of 25 shared species. Introductions of sports fish and for aquaculture were primarily responsible and extirpations were few, a common finding with fish introductions (Gido and Brown 1999, Moyle and Marchetti 2006).

Few invasive species can match the impact of the zebra mussel Dreissena polymorpha in North America, which has become so abundant in some river and lake habitats that native species are directly imperiled and ecosystem function has been greatly affected. Rarely common in small streams except below the outlets of infested lakes, they can reach very high densities (100-10,000 m2) in larger rivers, where they have strong negative effects on native mollusks by fouling their shells and outcompeting them for food (Strayer 1999a). As zebra mussels continue their spread, the loss of many species of native mussels is a serious possibility. Their influence over ecosystem function is a result of very high filtering rates, in the range of 10-100% of the water column per day, such that feeding by zebra mussels rather than downstream export becomes the dominant fate of transported food particles (Strayer et al. 1999). First observed in the Hudson River in 1991, zebra mussels had spread to all of the lower river's freshwater regions by 1992, having reached a biomass that exceeded all other het-erotrophs in the freshwater tidal Hudson. Phyto-plankton and small zooplankton declined precipitously, as did benthic consumers including native bivalves. However, macroinvertebrate populations increased, apparently benefiting from pseudofeces as a food supply and the mussels themselves as structural habitat. Light penetration increased greatly due to the filtering of the water column, and soluble reactive phosphorus (P) increased, presumably due to the reduction in algal populations. There is also evidence that macrophytes increased greatly following the zebra mussel's invasion, representing a shift in productivity from the river to its vegetated shallows. These effects also extended to the fish assemblage: open-water fish species declined in zebra mussel-infested river sections whereas littoral fishes increased substantially (Strayer et al. 2004). Zebra mussels and a few other bivalves may rank amongst the most influential of invaders of freshwater ecosystems, diverting resources from the water column to the benthos and to vegetated shallow margins.

13.2.3 Pollution

Declines in water quality result from industrial, municipal, residential, and agricultural sources that generate a wide variety of contaminants. Point source pollution comes from a single source and is often delivered through a pipe from an industry or municipal wastewater treatment plant, whereas nonpoint source (NPS) pollution comes from diffuse sources, such as fertilizer runoff and acid rain. The former can more readily be monitored and regulated, and in many developed countries considerable progress has been made toward reducing industrial and municipal waste. This is not true everywhere, of course. In China an estimated 80% of rivers and of drinking water supplies are reported to be polluted due to that country's recent and rapid economic development (Wang 2004). NPS pollution is much more difficult to manage, and in the United States is considered a more serious problem (USEPA 2000). Most NPS pollution consists of nutrients and sediments and is attributed to agriculture, but timber harvest, road building, and suburban sprawl all contribute as well.

Declining water quality due to many different pollutants unquestionably is an important aspect of the deterioration of aquatic ecosystems. Often the effects are local, but contamination of a river reach can block fish passage as effectively as a dam. Biodiversity is most likely to be threatened when the affected area is critical habitat to an already rare and endangered population, as with some fishes of the Yangtse River (Dudgeon et al. 2006), or when a large area is contaminated, as in the case of freshwater acidification and the reportedly widespread contamination of China's rivers.

13.2.3.1 Point source contaminants

Many North American rivers, although gradually improving, have experienced some degree of pollution and show regional variation in the types and sources of contaminants. In the Delaware River near Philadelphia, significant pollution was noted as early as 1799, low levels of dissolved oxygen were reported in 1915, and workers at river docks during World War II complained about the stench (Jackson et al. 2005). Sediments of the Connecticut River are burdened with a complex brew of metals and organic compounds including polychlorinated biphenyls (PCBs) and polycyclic hydrocarbons (PAHs), resulting in fish and shellfish advisories; in the middle Hudson River, commercial and sports fisheries are restricted due to PCB contamination.

The important point source discharges in urban areas are from wastewater treatment plants (WWTPs) and industry (Paul and Meyer 2001). Improvements in WWTP technology have resulted in significant reductions in chemical constituents in many countries. However, in many older cities, storm and sewer drains are combined in a single system, and so during high flows the volume of water exceeds the capacity of the WWTP, forcing it to bypass untreated wastewater directly into receiving waterways. Combined sewer overflows are reported from one fourth to one third of cities in the United States and United Kingdom, and the problem is further exacerbated by aging and leaking sewer systems and the proliferation of septic systems in exurban areas.

13.2.3.2 Runofffrom the land

Contaminants transported in urban and agricultural runoff are primary sources of nonpoint pollution to waterways, transporting sediments, nutrients, agricultural pesticides and herbicides, and various harmful substances (Table 13.2). More than one third of the river miles in the

United States are officially listed as impaired or polluted, and various categories of NPS pollution are considered leading causes (EPA 2002). Agricultural sources are responsible for 46% of the sediment, 47% of total P, and 52% of total nitrogen (N) discharged into waterways within the United States (Gianessi et al. 1986). Sedimentation affects the distribution of fish species, which vary widely in their tolerance for silted conditions. Agriculture increases nutrient levels due to fertilizers and animal waste, and also by increasing soil erosion, which particularly affects the transport of phosphorus. Urban areas can also be significant nutrient sources due to municipal wastes and fertilizers. In the Salt Fork River, an agricultural catchment in east-central Illinois that includes two substantial urban areas, urbanization was at least as important as agriculture in controlling instream nutrient concentrations (Osborne and Wiley 1988). This was true for soluble reactive P throughout the year and for nitrate-N during half of the year. During winter and spring, N fertilization of agricultural fields was the main determinant of instream concentrations.

Mass balance studies (Boyer et al. 2002) and analysis of fertilizer usage over time (Goolsby et al. 2001) show clearly that fertilizer usage is the primary driver of nutrient concentrations in rivers draining agricultural landscapes (Section 11.4). This is apparent from trends in nutrient concentrations in rivers (Figure 1311) and in sources of N inputs (Figure 13.12). Nitrate concentrations in the Mississippi River doubled between the 1950s and 1980s, coincident with steady growth in the application of fertilizers over the same period (Turner and Rabalais 1991). The Mississippi's greatly increased export of nitrate is responsible for a substantial area of low oxygen within the Gulf of Mexico, the result of enhanced algal productivity and decay (Raba-lais et al. 2001), and for changes to food web structure, owing to altered nutrient ratios (Turner et al. 1998). This is occurring in a region that supports 25% of US fish landings, and so has serious consequences for the regional fishing economy. Export of dissolved inorganic N by rivers is expected to continue to increase worldwide, due to increased fertilizer use and atmospheric deposition. Kroeze and Seitzinger (1998) forecast that by 2050, 90% of the DIN load of the world's rivers will be anthropogenic in origin.

A gradient of increasing urbanization has numerous impacts on stream ecosystems, including flashier flows from impervious surfaces and storm drains, nutrient and organic matter enrichment from lawn fertilizers and pet waste, altered channel morphology, and increased delivery of sediments and various toxins (Paul and Meyer 2001). Water entering streams can be significantly warmed during its passage over paved surfaces, and new construction is an important source of sediments. As a consequence of what has become known as the "urban stream syndrome'' (Walsh et al. 2005), biotic richness is reduced while dominance by tolerant species is

FIGURE 13.11 Historical trends in (a) phosphate and (b) nitrate concentrations in the River Thames, UK (Reproduced from Heathwaite et al. 1996.)

Was this article helpful?

0 0

Post a comment