Phosphorus and eutrophication of aquatic systems

Phosphorus and Eutrophicaation of Aquatic Systems

Phosphorus (P) is an essential element for all life forms. It is a mineral nutrient. Orthophosphate is the only form of P. that autotrophs are able to assimilate. Extracellular enzymes hydrolyze organic forms of P. To phosphate. Eutrophication is the overenrichment of receiving aquatic systems with mineral nutrients. The results are excessive production of autotrophs, especially algae and cyanobacteria. This high productivity leads to high bacterial populations and high respiration rates, leading to low oxygen concentrations or anoxia in poorly mixed bottom waters and at night in surface waters during calm, warm conditions. Low dissolved oxygen causes the loss of aquatic animals and release of many materials normally bound to bottom sediments including various forms of P. This release of P. reinforces the eutrophication.

Excessive concentrations of P. is the most common cause of eutrophication in freshwater lakes, reservoirs, streams, and headwaters of estuarine systems. In the ocean, N becomes the key mineral nutrient controlling primary production. Estuaries and continental shelf waters are a transition zone, where excessive P. And N. create problems. It is best to measure and regulate total P. inputs to whole aquatic ecosystems, but for an easy assay it is best to measure total P. concentrations, including paniculate P, in surface waters or N/P atomic ratios in phytoplankton.

Characteristics of Phosphorus

Phosphorus is a needed component of nucleic acids and many intermediary metabolites, such as sugar phosphates and adenosine phosphates, which are an important part of the metabolism of all life forms. With the exception of trace emissions of phosphines from volcanoes, the P. compounds found on the surface of the Earth are not volatile and transport through the atmosphere is primarily in dust or aerosols. Atmospheric flux rates are slow compared with those in surface waters. With few exceptions surface waters receive most of their P. In surface flows rather than in groundwater, since phosphates bind to most soils and sediments. The exceptions are where watersheds are of volcanic origin or where soils are water-logged and anoxic. Phosphorus only occurs in the pentavalent form in aquatic systems. Examples are orthophosphate, pyrophosphate, longer-chain polyphosphates, organic phosphate esters and phosphodiesters, and organic phosphonates. Phosphorus is delivered to aquatic systems as a mixture of dissolved and particulate inputs, each of which is a complex mixture of these different molecular forms of pentavalent

However P. is a very dynamic, biologically active element. After these P. inputs arrive in a receiving aquatic systems, the particulates may release phosphate and organic phosphates to solution in the water column and various P. compounds may be chemically or enzymatically hydrolyzed to orthophosphate, which is the only form of P. that can be assimilated by bacteria, algae, and plants. Particulates may be deposited in the bottom sediments, where microbial communities gradually use many of the organic constituents of the sediments, ultimately releasing much of their P. contents back to the water column as orthophosphate. Hence, one should not assume that particulate P. Or dissolved organic P. are inert in these aquatic systems because under appropriate conditions these forms of P. can be converted to dissolved orthophosphate.

Once delivered to a lake, reservoir, or estuary, P is usually kept fairly efficiently by a combination of biological assimilation and the deposition of sediments and biota to the bottom sediments. This efficient trapping of P. inputs makes these systems sensitive to pollution with excessive amounts of P. If the system is oligotrophic (low primary production), the bottom waters will have oxygen throughout the year and most of this P. will be stored in the bottom sediments.

However, in eutrophic systems (excessive primary production), Bottom waters often become anoxic during the growing season and even shallow waters may become diurnally anoxic at night during warm, windless weather. When these conditions occur, much of this P. In bottom sediments is released and diffuses back into the water column.

Evidence of the Key Role of Phosphorus

Over time ecologists developed the concept that plant and bacterial growth in an aquatic system would ultimately become limited by the availability of an essential element. This would then constitute the limiting nutrient for that system at that time, and inputs of that nutrient could be managed to limit eutrophication. The term limiting nutrient has been used in somewhat different ways, sometimes meaning limiting the growth of the present population, sometimes the limitation of growth over time with species composition changes, sometimes limiting the ultimate primary or net production of an ecosystem. Here the ultimate limitation of ecosystem primary production is inferred.

The diatom Cyclotella nana, grown in P-limited chemostats could only reach biomass atomic ratios of C. To P. Of 480 and N. To P. Of 35. This diatom had reached its limits of growth with the available P. In a series of bioassays of lake waters from the Great Lakes region of the USAu sing the Provisional Algal Assay Procedure (USDA1, 969), Selenastrum capricornutum cell number was found most often to respond to the addition of phosphate, rather than N, indicating that most of these lake waters contained limiting concentrations of P. Mesocosmex periments in which 320 L. Of Minnesota or Oregon lake water were enclosed in clear plastic bags and then enriched with various nutrients, found that P. was the primary controlling nutrient when positive responses were found (Powers et al., 1972). Mesocosm experiments in which 1000 to 4000 L. Of water from Lake Michigan were enclosed in clear plastic bags found that when P. was added, silica was reduced to levels that limited algal growth but N. was not (Schelske and Stoermer, 1972). They concluded that P. was the limiting nutrient in Lake Michigan, but that silica was becoming limiting for diatoms.

Somewhat later Lean and co-workers introduced the concept of an "index of P. deficiency." They used radioactive tracers to measure the turnover times of dissolved orthophosphate in lake surface waters. High turnover rates (short turnover times) indicated more P. limitation. This was further developed by measuring the ratio of C. fixation to phosphate uptake under various conditions. Atomic ratios of C. fixation to phosphate uptake varied from 1.2 to 206 depending on the degree of P. deficiency prior to the measurement. If the algae had previously been highly P. limited, they would fix a higher amount of C. per P. fixed.

A more direct measure of the key importance of P. In lake eutrophication was the work at the Experimental Lakes research area in northwestern Ontario. Whole lakes were enriched with P. For a period of years. These P-enriched lakes used atmospheric N. And C. For algal production and this resulted in significant increases in ecosystem primary production. Phosphorus additions triggered undesirable cyanobacterial blooms unless N. was also added. However, if C. Or N. were added, in the absence of P. enrichment the effects were minor ( Schindler, 1974, 1975, 1977). In another "whole lake" experiment, Lake Washington near Seattle, WA, had been heavily loaded with nutrients in sewage outfalls for many years and had become severely eutrophic.

Then, in 1963 the sewage effluent was diverted away from the lake. By 1969, chlorophyll in the summer and phosphate in the winter had declined to only 28% of previous years, but nitrate declined by only 10 to 20%. Lake Washington returned to a mesotrophic status. This was interpreted to mean that P. was the key limiting nutrient (Edmondson, 1970). Another example is Lake Erie that began experiencing dissolved oxygen depletion due to eutrophication. In 1968, its annual P. input was estimated to be 20-000 t and surface waters had an average of 22 ~zg of total P/L. By 1982, improved wastewater treatment had reduced annual P. inputs to 11-000 t and surface waters averaged only 12 I~g of total P/L (Boyce et al., 1987).

Many years of research on the effects of nutrient additions on lake productivity have led to a simple model that related algal biomass (Cla in mg/m3) t2o total P. input rates (Lp in g/m d-l), mean water depth (z in m), and outflow per unit l "a5k]e surface area (Qs in m/acre); Cla = (Lv/Qs)/[1 + (z/Qs) (according to Vollenweider model) This model fit the data from most the lakes and reservoirs that had been studied in the world and accurately predicted trophic status based only on input rates of one nutrient (P). This was very strong support for the importance of P. In the eutrophication of lakes. The Vollenweider model is still widely used by lake water quality managers, partly because of its simplicity. More sophisticated models often require more data than is available for a given lake.

If we accept that P. is usually the limiting nutrient in lakes, the next question is what is the relationship between P. enrichment and primary productivity. Prairie et al. analyzed data from 133 lakes for overall chlorophyll a relationships with total N. And total P. In the surface waters. The 133 lakes were selected from a larger set so that there was an even distribution of lakes that had ratios of total N. To total P. In surface waters, which varied from about 5 to 75. The LOWESS (a locally weighted regression) best fits for chlorophyll a concentrations in the various lakes were different functions for total P. And total N. LOWESS regression slopes and intercepts shifted with changing N/P atomic ratios with slopes maximized and intercepts minimized at an N/P ratio of 22. Similar analyses for data from 1041 lakes, not selected for N/P ratios, found that the log of chlorophyll a vs. log of summer total P. concentrations in surface waters was a sigmoid relationship that tended to flatten out at very high P. concentrations, rather than -- the linear one often assumed in the literature. It is apparent, therefore, that if a lake is already highly enriched with P, then adding more will have little effect, while adding N. will bring about major additional eutrophication.

Thus, for lakes it seems quite clear that P. is the nutrient most likely to be potentially limiting. But can we make the same statement about streams and rivers, reservoirs, or estuaries and coastal waters? Certainly, for estuaries and coastal waters the situation with respect to P-limitation of primary production is different. One of the first papers to conclude that there is a shift from P. To N. limitation as we move from fresh water to coastal waters was Ryther and Dunstan (1971).

Their view has become widely accepted (e.g., review by Nixon, 1981). However, Hecky and Kilham (1988) have challenged the basis for this conclusion. They felt that the generality and severity of N. limitation in" the oceans had not been rigorously established. Most scientists have put their efforts into determining why this apparent shift from P. limitation to N. limitation occurs.

Some of the more obvious reasons are the widely observed more efficient recycling of P. In estuaries, the high losses of fixed N. To the atmosphere due to denitrification in coastal waters (Nixon, 1981), and the role of sulfate in recycling P. In coastal sediments (Caraco et al., 1989). They found that a strong positive correlation existed between primary production and sulfate concentration in lakes and estuaries. The increases in primary production with increased sulfate concentrate had a higher slope in systems with anaerobic sediments, such as most estuaries. Under these conditions some of the sulfate is reduced to sulfides, which might bind the ferrous ions that are also produced in anaerobic sediments, preventing the ferrous ions from diffusing to the sediment/water column interface. In less reduced sediments a layer of oxidized sediments at the surface of the bottom sediments, coated with ferric hydroxide, is believed to form a barrier that traps diffusing phosphate before it can reach the overlying water column.

Regenerated phosphate is sufficient in Delaware Bay to supply almost all of the plankton P. demand except during the spring bloom (Lebo and Sharp, 1992). Estuaries, especially at their upstream ends, are transition zones. Sometimes they are P. limited in the spring, and N. limited in the summer and fall (Fisher et al., 1992; Lee et al., 1996). If lakes are primarily P. limited, the oceans ale primarily N. limited, and estuaries are transition zones, how about streams, rivers, and reservoirs? Although they are perhaps the least understood with respect to nutrient limitation, one might reasonably assume that they behave somewhat like lakes. However, unlike many lakes, unless they are highly enriched with nutrients they do not undergo anaerobic periods and thus are unlikely to release high concentrations of phosphate from bottom sediments. If they have long enough retention times, a given.volume of water moving downstream in a large river should behave much as though it were surface water in a lake or reservoir. Some differences include the "spiraling" of P. down the channel (Newbold et al., 1981; Elwood et al., 1983).

This is the result of uptake of P. By attached bacteria and algae (periphyton) and vascular plants and the binding of compounds in bottom sediments. When these P. compounds are released back into the water column, either from bottom sediments or attached biota, they move further down stream, before becoming attached again as the P. is cycled among the system components. Each such P. movement downstream in the water column is referred to as a "spiral." One of the earliest experimental studies of P. limitation in streams involved continuous addition for 8 d of diammoniump hosphate to a stream in Michigan. This resulted in an increase from

Background was about 4 ~g of P/L. The result was increased periphyton chlorophyll, higher rates of decomposition of leaf litter, and increased populations of snails and leaf-shredding macroinvertebrates. When phosphate was continuously added to a stream on the north slope of Alaska to increase the concentration in the stream by 10 ixg total P/L (Peterson et al., 1985), periphyton chlorophyll increased for 10 km downstream and the stream shifted from a heterotrophic to an autotrophic system. Effects ramified to increased bacterial activity and increases in the mean size of aquatic insects. These studies, although less numerous than was the case for lakes, strongly indicate that P. is also a key element controlling productivity of streams and rivers.

Are streams and rivers in the U.S.A. often highly polluted with P? The answer is yes. A trend analysis of 381 riverine sites in the USAfr om 1974 to 1981 (Smith et al., 1987) found that the average total P. concentration was 130 i.tg P/L, much higher than the levels attained in most of the fertilization experiments discussed above. Are these rivers improving with respect to P. concentrations? Fifty of these sites, mostly in the Great Lakes and upper Mississippi drainages, had declines in P. concentrations at 8.1% per year, mostly due to point source controls. Forty-three sites had increases at 7.4% per year, mostly due to increased diffuse sources of An interesting study by Soballe and Kimmel (1987) analyzed data from 345 streams from the National Stream Quality Accounting Network (NASOUAN) and 812 lakes and reservoirs from the National Eutrophication Survey (NES). A canonical discriminant analysis of algal cell abundance and nutrient status found that natural lakes and rivers formed end member populations, while reservoirs were intermediate and overlapping. Multiple regressions of algal cell abundance vs. total P. concentrations were significant, but different for all three categories of receiving water. Statistical models for each of the three types of water found that residence time, water depth, and water clarity were all important factors (r = 0.7. 0.6, and -0.4, respectively). Algal abundance per unit of increased from rivers to reservoirs to natural lakes. Thus, the effects of P. additions were most pronounced in lakes, primarily due to the long residence times typical of most lakes.People are attracted to lakes, rivers, and coastlines for diverse reasons. Clean water is a crucial resource for drinking, irrigation, industry, transportation, recreation, fishing, hunting, support of biodiversity, and sheer esthetic enjoyment. Throughout human history, water has been used to wash away and dilute pollutants. Pollutant inputs have increased in recent decades and have degraded water quality of many rivers, lakes, and coastal oceans. Degradation of these vital water resources can be measured as the loss of natural systems, their component species, and the amenities that they provide (U.S. EPA 1996, Postel and Carpenter 1997). Water shortages are increasingly common and likely to become more severe in the future (Postel et al. 1996, Postel 1997). Water shortage and poor water quality are linked, because contamination reduces the supply of water and increases the costs of treating water for use. Preventing pollution is among the most cost-effective means of increasing water supplies.

Eutrophication caused by excessive inputs of phosphorus (P) and nitrogen (N) is the most common impairment of surface waters in the United States (U.S. EPA 1990), with impairment measured as the area of surface water not suitable for designated uses such as drinking, irrigation, industry, recreation, or fishing. Eutrophication accounts for 50% of the impaired lake area and 60% of the impaired river reaches in the United States (U.S. EPA 1996), and is the most widespread pollution problem of U.S. estuaries (NRC 1993a). Other important causes of surface-water degradation are siltation caused by erosion from agriculture, logging and construction (which also contribute to eutrophication), acidification from atmospheric sources and mine drainage, contamination by toxins, introduction of exotic species, and hydrologic changes (NRC 1992).

Chemical inputs to rivers, lakes, and oceans are classified as point or nonpoint sources (Table 1). Pollutant discharges from point sources such as municipal sewage treatment plants tend to be continuous, with little variability over time. Often they can be monitored by measuring discharge and chemical concentrations periodically at a single place. Consequently, point sources are relatively simple to measure and regulate, and can often be controlled by treatment at the source. Nonpoint inputs can also be continuous, but are more often intermittent and linked to seasonal agricultural activity or irregular events, such as heavy precipitation or major construction. Nonpoint inputs often derive from extensive areas of land and are transported overland, underground, or through the atmosphere to receiving waters. Consequently, nonpoint sources are difficult to measure and regulate. Control of nonpoint pollution centers on land management practices and control of release of pollutants to the atmosphere, and may affect the daily activities of millions of people.

Nonpoint inputs are the major source of water pollution in the United States (U.S. EPA 1990, 1996). The National Water Quality Inventory stated that "the more we look, the more we find" (U.S. EPA 1988). For example, 72 -- 82% of eutrophic lakes would require control of nonpoint phosphorus inputs to meet water quality standards, even if point inputs were reduced to zero (Gakstatter et al. 1978).

In many cases, point sources of water pollution have been reduced, owing to their relative ease of identification and control. Point sources are still substantial in some parts of the world, and may increase with future expansion of urban areas and aquaculture. This report focuses on nonpoint sources, not because point sources are unimportant, but because nonpoint inputs are often overlooked. In addition, (1) restoration of most eutrophic waters requires the reduction of nonpoint inputs of P. And N; (2) we have a sound scientific understanding of the causes of nonpoint nutrient pollution and, in many cases, we have the technical knowledge needed to decrease nonpoint pollution to levels compatible with water quality standards; and (3) the most important barriers to control of nonpoint nutrient pollution appear to be social, political, and institutional. We hope that our summary of the scientific basis of the problem will inform and support the debate about solutions.

Why Is Nonpoint P. Pollution a Concern?

Direct health effects

Phosphorus in water is not considered to be directly toxic to humans and animals (Amdur et al. 1991). Because of this, no drinking water standards have been established for P (U.S. EPA 1990). Any toxicity caused by P. In freshwaters is indirect. The proximal cause is toxic algal blooms or anoxic conditions stimulated by P. pollution.

What Are the Sources of Nonpoint Pollution?

Nonpoint P. pollution is caused primarily by agricultural and urban activities (Novotny and Olem 1994, Sharpley et al. 1994). Atmospheric deposition from diverse sources can add significant amounts of P. To surface waters (Howarth et al. 1996). Agriculture is the predominant source of nonpoint nutrient pollution in the United States (NRC 1992, U.S. EPA 1996).

Agriculture

On the world's agricultural lands, nutrient transport by farming systems has overwhelmed natural nutrient cycles (Fig. 1). Globally, more nutrients are added as fertilizers than are removed as produce. Fertilizers are moved from areas of manufacture to areas of crop production. They are partly incorporated into crops, which are then moved to localized areas of human consumption and livestock production. Thus, there is a net transport of P. from sites of fertilizer manufacture to sites of fertilizer deposition and manure production (Beaton et al. 1995). This flux creates a nutrient surplus on agricultural lands, the underlying cause of nonpoint pollution from agriculture.

Fertilizer. -- Phosphorus is accumulating in the world's agricultural soils. Between 1950 and 1995, 600 x 106 Mg of fertilizer P. were applied to Earth's surface, primarily on croplands (Brown et al. 1997, FAO 1950 -- 1995). During the same time period, 250 x 106 Mg of P. were removed from croplands through harvest (Beaton et al. 1995, Brown et al. 1997, FAO 1950 -- 1995). Some of the harvested P (50 x 106 Mg) was reapplied to cropland as animal manure (NRC 1993b). Thus, the net addition of P. To croplands over this period was 400 x 106 Mg. This applied P. may either remain in soils or be exported to surface waters by erosion or leaching. The majority of applied P. remains on croplands, with only 3 -- 20% leaving by export to surface waters (Caraco 1995). It is likely, therefore, that 350 x 106 Mg of P. have accumulated in the world's croplands. The standing stock of P. In the upper 10 cm of soil in the world's croplands is 1300 Mg (Pierrou 1975). Therefore, the net addition of 350 x 106 Mg between 1950 and 1995 would have increased the P. content of agricultural soils by 25%. In the United States and Europe, only 30% of the P. input in fertilizers is output in produce, resulting in an average accumulation rate of 22 kg ha?1 y?1 for surplus P (Table 4). At the watershed scale, excess inputs of P. To agriculture relative to outputs in produce are closely linked to eutrophication of surface waters (Fluck et al. 1992).

Manure. -- Intensive animal production generally involves feeding large numbers of animals in small areas (NRC 1993b). For example, only 4% of the cattle feedlots in the United States produce 84% of the cattle (NRC 1993b). These large concentrations of animals create enormous amounts of waste. Disposal problems are comparable to those for raw human sewage, but the regulatory standards for animal waste are generally far less stringent than those for human sewage.

Nutrients in manure can be recycled by applying the manure to cropland. However, manure yields from concentrated livestock operations often exceed the capacity of croplands to sequester the nutrients (NRC 1993b). At typical stocking rates for feedlots, an area of cropland 1000 times greater than the feedlot area is required to distribute manure nutrients at levels similar to crop demand (NRC 1993b). This much land may not be available, so manure is applied to excess. These nutrients build up in soil, run off or infiltrate to water supplies, or (in the case of N) can enter the atmosphere.

Transport to aquatic ecosystems. -- Increased fluxes of P. To surface waters have been measured after application of fertilizer or manure to farm land (Sharpley and Rekolainen 1996). Fertilizer P. losses in runoff are generally

The amount of P. lost to surface waters increases with the P. content of the soil (Fig. 2). Relationships similar to those in Fig. 2 have been demonstrated for a diversity of soils (Sharpley et al. 1996). In the long-term, this particulate P. can become available to the aquatic biota (NRC 1992, Sharpley et al. 1994).

In recent decades, N transport to the oceans has increased (Howarth et al. 1996). It is correlated with various indices of human activity in the watershed (Cole et al. 1993, Howarth et al. 1996, Vitousek et al. 1997). Similarly, P flux to the oceans in rivers is positively correlated with human population density (Caraco 1995). Globally, P flux to coastal oceans has increased from an estimated pristine flux rate of 8 x 106 Mg/yr to the current rate of 22 x 106 Mg/yr (Howarth et al. 1995). Of this increase, 30% is attributed to P. enrichment of agricultural soils and the remainder to increasing rates of erosion.

Urban runoff

A significant amount of P. And N. enters surface waters from urban nonpoint sources, such as construction sites, runoff of lawn fertilizers and pet wastes, and inputs from unsewered developments. Urban runoff is the third most important cause of lake deterioration in the United States (U.S. EPA 1990), affecting 28% of the lake area that does not meet water quality standards. Urban point sources of water pollution, such as sewage and industrial discharges, are also significant and often are managed intensively.

Construction sites are a critical area of concern for urban nonpoint pollution. Although construction sites may occupy a relatively small percentage of the land area, erosion rates for construction sites can be extremely high, so the total nonpoint pollution yield is large. Erosion rates from watersheds under development approach 50000 Mg km?2 yr?1, compared to 1000 -- 4000 Mg km?2 yr?1 for agriculture and

What Can Be Done about It?

Unless current practices are modified, nonpoint pollution of surface waters will increase in the future. Some factors that drive this expectation are the substantial and growing accretion of P. In agricultural ecosystems; a growing human population; the tendency for people to eat diets rich in meat; increasing livestock production; growth of urban areas, with associated development and erosion. However, this pessimistic forecast could prove to be incorrect, because there are a number of ways that nonpoint pollution can be reduced.

Landscape management

Riparian (shoreline) vegetation can significantly reduce nonpoint nutrient flows to surface waters. This vegetation also makes important contributions to fish and wildlife habitat and regional biodiversity. Interest in the use of riparian vegetation for nonpoint pollution control has grown rapidly in recent years, as evidenced by more than 500 publications on the subject.

Agricultural P. management

The ultimate causes of nonpoint pollution from agricultural lands are excessive fertilizer use and high-density livestock operations. There are direct solutions. Fertilizer application can be reduced to match crop needs. Wastes from high-density livestock operations can be managed as a point source of pollution. Nutrients in manure can be used as fertilizer, or nutrients can be removed (as in sewage treatment) before wastes are discharged to surface waters. Work to implement these solutions now focuses on establishing threshold levels of soil nutrients that threaten water quality, delineation of intensive sources of pollutants, and controlling both nutrient sources and transport.

Thresholds. -- Threshold levels of soil nutrients that create unacceptable threats to water quality are needed to provide a firm basis for regulations that protect aquatic resources. Defining thresholds has been controversial, in part because data are insufficient. Unfortunately, the database relating soil nutrient concentrations to runoff is limited to a few soils and crops, making it difficult to extrapolate these data to other regions. Because costs of manure management are significant, those most affected by thresholds have vigorously challenged their scientific basis. A stronger scientific foundation can, and should, be developed for soil nutrient thresholds so that scientifically-based standards can be promulgated and defended.

Source area delineation. -- Typically, >90% of the P. export from watersheds derives from

Source management. -- P runoff can be greatly reduced if nutrients are applied at rates that match their uptake by crops, and if fertilizers are applied when crops are growing rapidly (Sharpley et al. 1994). Also, dietary P. inputs to livestock can be matched to the animals' requirements to decrease the amounts of P. excreted on to land. Source management can significantly reduce concentrations of P. In the runoff entering aquatic ecosystems. Aggressive treatment of dairy wastes in Florida, for example, reduced total P. concentrations in surface water by 62 -- 87%.

Transport management. -- Transport of P. To surface waters by erosion and runoff may be reduced by riparian zones, buffer strips, conservation tillage, terracing, contour tillage, cover crops, and retention ponds. Vegetated buffer strips in riparian zones, for example, reduce P. transport to streams by 50 -- 85%. Without concurrent reductions in nutrient sources to soils, however, soil nutrients will continue to accumulate.

Control of urban runoff

Control of urban nonpoint pollution is a well-developed branch of civil engineering, with an extensive and sophisticated literature. Optimization of sewer systems is a key goal. Other approaches used to varying degrees include the creation of retention ponds, wetlands, and greenways as an integrated component of urban stormwater management systems; litter control and street sweeping; reduction of impervious areas; and reduction of erosion, especially from construction sites.

Conclusion

A sound fundamental understanding of the processes that cause nonpoint phosphorus pollution and eutrophication exists. At regional and global scales, the causes and consequences of nonpoint pollution are clear. There is need for site-specific analyses, and the capacity for such analyses is well developed and improving. The most critical need in reducing nonpoint pollution and mitigating eutrophication may be creative institutional mechanisms that match scientific understanding with social realities (Gunderson et al. 1995). Although science alone cannot solve the problem, our evaluation of the literature suggests that the necessary scientific understanding is well developed and could be readily mobilized in the search for solutions.

Baines, S. B, Pace, M.L., and Karl, D.M. 1994. Why does the relationship between sinking flux and planktonic primary production differ between lakes and oceans? Limnol. Oceanogr. 39: 213-226.

Downing, J.A. 1997. Marine nitrogen: phosphorus stoichiometry and the global N: P. cycle. Biogeochemistry 37: 237-252.

Downing, J.A., Osenberg, C.W., AND Sarnelle, O. 1999. Meta-analysis of marine nutrient-enrichment experiments: Variation in the magnitude of nutrient limitation. Ecology 80: 1157-1167.

Guildfords., J., and R.E. Fisher, R.R., and others. 1999. Spatial and temporal variation of resource limitation in Chesapeake Bay. Mar. Biol. 133: 763-778.

Fisher, T.R., Hagy III, J.D., Boynton, W.R., and Williams, M.R. 2006. Cultural Eutrophication in the Choptank and Patuxent Estuaries of Chesapeake Bay. Limnology and Oceanography, Vol. 51, No. 1, Part 2: Eutrophication of Freshwaterand Marine Ecosystems, pp. 435-447

Hecky. 2000. Total nitrogen, total phosphorus, and nutrient limitation in lakes and oceans: Is there a common relationship? Limnol. Oceanogr. 45: 1213-1223.

Hecky, R.E. And P. Kilham 1988. Nutrient limitation of phytoplankton in freshwater and marine environments: A review of recent evidence on the effects of enrichment. Limnol. Oceangr. 33: 796-822.

Howarth, R.W. And Marino, R. 2006. Nitrogen as the Limiting Nutrient for Eutrophication in Coastal Marine Ecosystems: Evolving Views over Three Decades. Limnology and Oceanography, Vol. 51, No. 1, Part 2: Eutrophication of Freshwaterand Marine Ecosystems (Jan., 2006), pp. 364-376.

Hoyer, M.V., T.K. Frazer, S.K. Notestein, and D.E. Canfield. 2002. Nutrient, chlorophyll, and water clarity relationships in Florida's nearshore coastal waters with comparisons to freshwater lakes. Can. J. Fish. Aquat. Sci. 59: 1024-1031

Hutchinson., E. 1973. Eutrophication. Am. Sci. 61: 269-279.

, and V.T. Bowen. 1947. A direct demonstration of the phosphorus cycle in a small lake. Proc. Natl. Acad. Sci. USA 33: 148-153.

Likens, G.E. [Ed.]. 1972. Nutrients and eutrophication. Limnol. Oceanogr. Special Symp 1: American Society of Limnology and Oceanography.

National Research Council. 2000. Clean coastal waters: Understanding and reducing the effects of nutrient pollution. National Academy Press

Paerl, H.W., Valdes, L.M., Peierls, B.L., Adolf, J.E. And Harding, LW. 2006. Anthropogenic and Climatic Influences on the Eutrophication of Large Estuarine Ecosystems. Limnology and Oceanography, Vol. 51, No. 1, Part 2: Eutrophication of Freshwaterand Marine Ecosystems, pp. 448-462

Redfield, A.C. 1958. The biological control of chemical factors in the environment. Am. Sci. 46: 205-222.

Redfield, A.C. 1934. On the proportions of organic derivatives in sea water and their relation to the composition of plankton, p. 176-192. In James Livingstone Memorial Volume. Univ. Of Liverpool.

Schindler, D.W. 2006. Recent Advances in the Understanding and Management of Eutrophication. Limnology and Oceanography, Vol. 51, No. 1, Part 2: Eutrophication of Freshwaterand Marine Ecosystems, pp. 356-363.

Smith, V.H. 2006. Responses of Estuarine and Coastal Marine Phytoplankton to Nitrogen and Phosphorus Enrichment: Limnology and Oceanography, Vol. 51, No. 1, Part 2: Eutrophication of Freshwater and Marine Ecosystems, pp. 377-384.

Weber, C.A. 1907. Aufbau und Vegetation der Moore Norddeutschlands. Beibl. Bot. Jahrb. 90: 19-34.