Ecosystem Modeling in Cobscook Bay, Maine: A Boreal, Macrotidal Estuary 2004 Northeastern Naturalist 11(Special Issue 2):87–100 Predicted Nutrient Enrichment by Salmon Aquaculture and Potential for Effects in Cobscook Bay, Maine JOHN W. SOWLES 1,* AND LAURICE CHURCHILL 1 Abstract - Salmon aquaculture has been a prominent feature of Cobscook Bay since the late 1980s, providing jobs in an economically depressed area of the State of Maine. Rearing finfish in moored floating pens is not without environmental consequences, however, with waste feed, feces, and dissolved nutrients discharged directly to surrounding waters. Near-field benthic effects have been well studied in Cobscook Bay, but far-field effects of nutrient enrichment from salmon aquaculture have not. Via two independent indirect methods, this paper estimates the amount of nitrogen and phosphorus discharged by salmon farms in Cobscook Bay. Results suggest that between 1995 and 1996, during the Cobscook Bay research program, salmon aquaculture contributed an annual load of about 360 metric tons of nitrogen and 85 metric tons phosphorus to Cobscook Bay. Compared to the already high nutrient flux from sources outside Cobscook Bay, we conclude it is unlikely the incremental contribution from aquaculture is measurably enhancing planktonic primary production. However, we acknowledge the need for further study on benthic macroalgae. Introduction Sheltered from strong winds and waves, Cobcook Bay’s 5–7 m tides and strong currents ensure a liberal supply of clean, oxygen-rich water (Brooks et al. 1999). Because water temperatures in Cobscook Bay are moderated by incoming deeper shelf water via the Gulf of Maine, temperatures seldom fall below 4 oC in winter or exceed 14 oC in summer. High flushing also promotes assimilation and dispersion of metabolic wastes. Combined, these conditions make Cobscook Bay an ideal environment for the culture of coldwater marine finfish such as Atlantic salmon (Salmo salar L.) Salmon farming in Cobscook Bay began on a small scale in the early 1980s in wooden pens made by local fishermen. By the early 1990s, farms were a mix of small family “experiments” and larger corporate operations. By 1996, production had grown to nearly 6 million kg per year (Fig. 1). The industry contributed about $30 million annually to the economy of downeast Maine and provided nearly 700 direct and indirect jobs (Young et al. 1998). By the late 1990s, Cobscook Bay salmon aquaculture was in a highly competitive global market. As family owned and 1Maine Department of Marine Resources, PO Box 8, West Boothbay Harbor, ME 04575. Corresponding author - john.sowles@maine.gov. 88 Northeastern Naturalist Vol. 11, Special Issue 2 operated farms became less economically viable, leases were conveyed to larger operators with the financial resources to compete on an international level. Today, industry activity is primarily composed of three large companies. In 2001, a viral outbreak of infectious salmon anemia (ISA) resulted in high mortality of farmed salmon in Cobscook Bay. To control the outbreak, an eradication order was issued to depopulate the Bay of farmed fish. After several months of fallowing to break the disease cycle, the Bay was restocked in late spring and early summer of 2002. Today, Cobscook Bay contains about half the number of farmed fish held in the 1990s, a level that is likely to continue for the forseeable future as one part of a disease management strategy. The marine phase of salmon aquaculture begins when young salmon that are reared in freshwater hatcheries are transferred to floating net pens. Each pen has a fine mesh net about 10 m deep to contain the fish. Surrounding the containment net is a heavier, larger mesh underwater predator net to deter seals. A fine mesh net is placed over the entire pen to prevent entry by avian predators. Pens may contain from 15,000 to over 50,000 fish, depending on the age of fish, and the size and type of cage. Leases may contain upwards of 20 pens. Several times a day, fish are fed a diet specially formulated for Atlantic salmon depending on body weight, temperature, and fish behavior. As feed is the most expensive cost in salmon husbandry, farms are equipped with feed monitoring equipment such as underwater video cameras or acoustic devices used to prevent feed waste and subsequent loss to the environment. Sites are leased by the State to aquaculturists for 10-year periods after meeting approval criteria. Approval criteria require that a farm not unreasonably interfere with traditional fisheries, navigation, access to Figure 1. Farm raised salmon harvested from Cobscook Bay region and elsewhere in Maine from 1989 through 2003 (FAMP 1989–2003). 2004 J.W. Sowles and L. Churchill 89 and from a riparian owner’s property, ecologically significant species, other existing aquaculture operations, certain public facilities, and other existing uses (Title 12 Maine Revised Statutes Annotated §6072). As with any form of concentrated animal husbandry, aquaculture has the potential for adverse consequences. Organic enrichment of bottom sediments, nutrient enrichment of the water column, physical habitat alteration, contamination by therapeutants and pesticides, genetic and ecological interactions between native and farmed populations, and interference with wildlife (e.g., marine mammals and nesting birds) have been reported (Enell and Ackefors 1992, Goldberg and Triplett 1997, Gowen and Bradbury 1987, Wildish et al. 1993). Maine’s challenge has been to take advantage of aquaculture’s positive aspects while safeguarding public waters from harm. As a condition of the lease to protect environmental quality, aquaculturists are responsible for maintaining standards of the state Water Classification Program (Title 38 Maine Revised Statutes Annotated §464) that limit impacts to the water column, benthic habitat, and biological communities under and around the pens. From 1988 until 2004, compliance was monitored by the Department of Marine Resources’ Finfish Aquaculture Monitoring Program (FAMP) (Title 12 Maine Revised Statutes Annotated §6077) in which every stocked farm was inspected in the fall when salmon biomass, feeding, and water temperatures were at a peak. Divers videotaped a transect under and away from the pens, collected sediment cores for sampling benthic infauna, and water samples for analyzing the chemistry and vertical oxygen profiles of the water column. In addition, each operator reported the number of fish on site, biomass, and amount of feed used monthly to the Department of Marine Resources. In 2004, monitoring responsibility was transfered to the companies as part of a wastewater discharge license required by the Maine Department of Environmental Protection. Although near-field (within 100 m) benthic effects of netpen systems were emphasized in the early monitoring program, concern over cumulative far-field effects from aquaculture emerged in the late 1980s when salmon aquaculture began to experience rapid growth. Of particular concern to State water quality managers was the issue of cultural eutrophication. The innermost portion of Cobscook Bay was presumed to have slower flushing than the Outer Bay and thus be more vulnerable to nutrient enrichment. The Inner Bay was also well known for its rich invertebrate communities (Trott 2004). Together, these concerns led to legislation that prohibited any new wastewater discharges inside a line between Denbow and Leighton Necks (Fig. 2; Title 38 Maine Revised Statutes Annotated, Chapter 3 §469). Since finfish aquaculture involves the discharge of dissolved and solid material, about one third Cobscook Bay (Fig. 2) was excluded for use by finfish aquaculture. 90 Northeastern Naturalist Vol. 11, Special Issue 2 In 1995 and 1996, the years during which field work for the Cobscook Bay modeling project was conducted (Larsen 2004), 14 leases were stocked with fish (Fig. 2). This paper presents our approach to estimating cumulative nutrient loading to Cobscook Bay from finfish aquaculture for the period of the Cobscook Bay ecosystem modeling project and our assessment of potential adverse effects. Methods Obtaining empirical information on nutrient loading from finfish pen culture is impractical, if not impossible. Unlike land-based production Figure 2. Location of Maine salmon leases in Cobscook Bay region. Closed circles were active farms in 1995–1996. Open circles were inactive farms in 1995–1996. Shaded area of bay west of Denbow and Leighton Neck delineates area where discharges are prohibited. 2004 J.W. Sowles and L. Churchill 91 facilities that have a discharge pipe from which effluent samples may be collected, net pen wastes are released through the sides and bottom of the cages where they rapidly diffuse and disperse into the surrounding water. In high current environments like Cobscook Bay (Brooks 2004), this rapid dispersal makes it difficult to attribute water quality changes to a source, especially a diffuse source like a net pen. In the nearby Letang/Western Isles region of the Bay of Fundy, for example, Wildish et al. (1993) reported difficulty discerning water quality changes beyond only a few meters from cages. Salmon husbandry presents a unique challenge in that, unlike effluent from an industrial process or municipal wastewater treatment plant, loading varies with time of day, month, and year in response to husbandry, disease, water temperature, fish physiology, and, not to be underestimated, fish behaviour (Axler et al. 1997). To overcome some of these difficulties, we estimated nutrient loading to Cobscook Bay from salmon farming by aggregating monthly feed and harvest figures for all farms in the Bay. We used the period 1995 to 1996 to coincide with ambient nutrient data collected in Cobscook Bay during May, July, October, and November 1995 cruises (Garside et al. 2004, Phinney et al. 2004) as well as field work by Vadas et al. (2004a,b,c) in summer 1995 and 1996. This period also represented relatively stable conditions prior to the ISA outbreak, when salmon production in Cobscook Bay was near peak production for the decade (Fig. 1). Absent direct measurement, we estimated mass loading using two indirect yet independent methods. The first method employs a loading coefficient and the second method is a simple input-output mass balance model that includes a combination of measured and deduced values. Similar approaches have been employed by others (Ackefors and Enell 1990, 1994) and incorporated into aquaculture development planning (Gillibrand et al. 2002, Gowen 1994). Loading coefficient method The loading coefficient method applies a nutrient load per unit of fish produced using the following simple formula: L= Lc P where L = annual nutrient load in kg, Lc = the loading coefficient in kg of nutrient element per metric ton of production, and P= metric tons (live weight) fish produced in a year. Table 1. Loadings (kg) per metric ton (live wt.) production of Atlantic salmon. Ackesfor Cho and Solbe and Enell Beveridge Bureau Constituent 1988 1994 1996 1997 Lc Total nitrogen 67.5 63–93 56–76 30–50 65 Total phosphorus 15.7 6.5–12 5–11 4.3–7.3 10 92 Northeastern Naturalist Vol. 11, Special Issue 2 Annual production (P) was gathered from monthly harvest reports submitted to the Department of Marine Resources (DMR) by growers (FAMP 1995 and 1996). Loading coefficients (Lc) for nitrogen and phosphorus in Table 1 were derived from a range of literature values (Ackefors and Enell 1994, Beveridge 1996, Cho and Bureau 1997, Solbe 1988). Atlantic salmon diets have become progressively more efficient resulting in less nutrient loss to the environment (Beveridge 1996, Hardy 1999). This increase in efficiency is reflected in Table 1, which shows decreases in Lc over 10 years. To avoid underestimating loading, we chose to employ slightly higher than average loading coefficients of 65 kg total nitrogen and 10 kg total phosphorus. Input-output mass balance method The mass balance method accounts for inputs and outputs in more detail. By knowing weights and chemical composition of the various inputs (feed and smolts) and outputs (mortality removal and harvest), net loading to the environment may be estimated. Loading is calculated as the difference between material added to the system minus material removed. Proximal analysis of the feed showed that nitrogen and phosphorus comprised about 7.0% and 1.5% by fresh weight, respectively (Findlay and Watling 1994). Atlantic salmon contain about 3.25% nitrogen (Vinogradov 1953) and 0.44% phosphorus (Rottiers 1993) on a live weight basis. Smolt stocking and mortality numbers were available from the monthly reports submitted to the DMR (FAMP 1995 and 1996). Results Growers in Cobscook Bay reported using approximately 7,500,000 kg of feed and produced about 5,500,000 kg (live weight) of adult salmon during the Cobscook Bay project study period. This reflects a conversion efficiency of about 1 unit fresh weight to 1.4 units live weight salmon, a ratio commonly observed for Atlantic salmon husbandry in the mid 1990s (Naylor et al. 2000). Using the loading coefficients (Lc) from Table 1 and the input-output mass balance figures above, we estimated two independent annual nutrient loads to Cobscook Bay (Tables 2 and 3). Estimates of annual loading from the two methods are in relatively close agreement, with between 339,000 to 357,500 kg TN and 55,000 to 85,000 kg TP, but somewhat lower than an estimate of 438,000 kg N per Table 2. Estimated annual nutrient load (kg) to Cobscook Bay from salmon aquaculture using the loading coefficient method. Constituent Production x coefficient Annual load Total N 5500 MT x 65 kg/MT = 357,5000 kg TN Total P 5500 MT x 10 kg/MT = 55,000 kg TP 2004 J.W. Sowles and L. Churchill 93 year made by Campbell (2004), who used a similar input-output mass balance approach. The discrepancy may be explained, in part, by our use of slightly different feed nitrogen contents. Campbell used 7.68%, whereas we used 7.0% N content of feed based on actual chemical analyses (Findlay and Watling 1994). Loading on an annual or average basis, however, is probably not especially useful for assessing potential impacts as it ignores seasonality associated with both aquaculture husbandry and ecosystem processes. In fish culture, feeding, standing stock, and metabolism are at the annual maximum when seawater temperatures are highest and are at the annual minimal when water temperatures are coldest. Over a year, daily loading may vary by an order of magnitude or more. In Cobscook Bay, Table 3. Estimated annual nutrient load (kg) to Cobscook Bay from salmon aquaculture using the mass balance method Worksheet for nitrogen mass balance 7,500,000 kg feed x 0.07 N = 525,000 kg TN 1,500,000 smolts x 0.15 kg/smolt x 0.03N = 6750 kg TN 5,500,000 kg harvest x 0.03 N = -165,000 kg TN 950,000 kg mortality x 0.03 N = -28,500 kg TN Total N to Cobscook Bay = 338,250 kg TN Worksheet for phosphorus mass balance 7,500,000 kg feed x 0.015 P = 112,500 kg TP 1,500,000 smolts x 0.15 kg/smolt x 0.0044 P = 990 kg TP 5,500,000 kg harvest x 0.0044 P = -24,200 kg TP 950,000 kg mortalitiy x 0.0044 P = -4180 kg TP Total P to Cobscook Bay = 85,110 kg TP Figure 3. Monthly loading estimates to Cobscook Bay from salmon aquaculture derived from monthly monitoring reports (DMR files, 1995–1996). 94 Northeastern Naturalist Vol. 11, Special Issue 2 temperatures peak in late August–early September and are lowest in February. Although Garside and Garside (2004) concluded that nitrogen was not limiting in Cobscook Bay, we pursued a scenario in which adding nutrients in late summer, when primary production in the open ocean is considered nitrogen limited, could promote phytoplankton growth. To do so, we looked at nutrient loading over monthly increments to predict what impacts might occur if nutrients were to become limiting in Cobscook Bay (Fig. 3). To assess potential significance of these loads to Cobscook Bay, we constructed a scenario biased toward “worst case.” Brooks et al. (1999) found that residence times in the Outer Bay may be less than a day, while those in Whiting Bay, Dennys Bay, and several gyres (e.g., in South Bay) may be more on the order of 5–8 days. We asked the question, “what might be the maximum change in ambient nutrient concentration if the entire load for a 10-day period during the September peak feeding period was released instantaneously to the Bay?” For this exercise, we made the following assumptions:1) the entire 10-day load would be instantaneously introduced, 2) the load would be uniformly mixed throughout the Bay, and 3) all nitrogen and phosphorus would be immediately available for plant uptake in a dissolved form. To estimate change in ambient concentrations, we applied the following simple dilution formula: DC = L/V, where DC is change in average bay concentration, L = 10/30ths of the September nitrogen and phosphorus load from salmon aquaculture or 18,385 kg and 3064 kg, respectively, and V = high tide volume of Cobscook Bay or 1.05E+9 m3 (Brooks 2004). Units were then converted to micromoles (μM) to enable comparison with Garside and Garside (2004) and Phinney et al. (2004). Following this scenario, salmon aquaculture could contribute up to 1.25 μM Total N and 0.2 μM Total P to ambient levels at peak production in early fall. Discussion Significance to the Cobscook Bay ecosystem Certainly, an annual load of 360,000 kg total nitrogen and 85,000 kg total phosphorus is substantial. However, the important question is whether or not such a load produces undesireable ecological consequences. Potential risk from these nutrient additions may be discussed from two perspectives: exposure and effects. The two are very different; exposure is the amount of stressor added to a system, while effect is the manifestation of that stressor by the system. Context of exposure Data from an October 1995 research cruise (Garside et al. 2004) enable us to compare our September “worst case” scenario against ambient 2004 J.W. Sowles and L. Churchill 95 water quality. October dissolved inorganic nitrogen (DIN) and dissolved orthophosphorus (PO4-P) averaged 7.5 μM and 0.62 μM respectively over 168 ambient water column samples which compare to the 1.25 μM total nitrogen and 0.2 μM total phosphorus estimated by our methods. Thus, during peak production in late September, salmon aquaculture could contribute up to 17% and 32% of the nitrogen and phosphorus pool circulating in the Bay in late September, respectively. Additional context for nitrogen is provided by comparing Garside and Garside’s (2004) estimated daily tidal flux of 70 metric tons nitrate nitrogen to less than two metric tons (3%) total nitrogen delivered daily (1/30th September total nitrogen load [from Fig. 3]) by salmon farms in Cobscook Bay. Garside and Garside (2004) did not estimate a daily tidal flux on which to base a phosphorus comparison. Despite the arguable assumptions of our simple model, we believe our estimates for the relative contribution of nutrients from Maine salmon farms are biased toward “worst case” and that loading is actually somewhat less. For example, thus far we have discussed nitrogen and phosphorus as total nitrogen and total phosphorus, yet our comparisons are against dissolved forms of nitrogen and phosphorus measured in the Cobscook Bay modeling study. Not all nitrogen and phosphorus lost by fish is immediately available as dissolved nitrogen. Ackefors and Enell (1990) report that about 78% of the nitrogen and 23% of phosphorus is dissolved. Applying these percentages, salmon farming’s peak contribution to the dissolved pool of nitrogen and phosphorus in Cobscook Bay would be 1 μM and 0.05μM respectively or 13% of the dissolved nitrogen and 8% of the dissolved phosphorus. At other times of the year, the absolute and relative contribution of nutrients from salmon farming would be much less, especially in late winter–early spring when feeding and metabolism are at a minimum. Furthermore, since most of the farms, and hence load, lie in the Outer Bay where Brooks (2004) estimates a 1–2 day retention, using a 10- day aggregate load is also an exaggeration. Potential for adverse effects Regardless of exposure level, effects from even seemingly small increases of a stressor may have adverse consequences if the system is approaching a critical level. Conversely, a large increment of change may not necessarily result in an adverse impact if an even larger increment of change is needed before reaching a critical point or if factors other than nutrients control productivity. Here we are concerned with the effects of eutrophication: algae blooms, algal mats, and hypoxia. Garside and Garside (2004) and Phinney et al. (2004) conclude that phytoplankton in Cobscook Bay is not nutrient limited, but rather light limited. The latter authors describe Cobscook Bay as a “high nutrient/low chlorophyll” (HNLC) system where chlorophyll concentrations are lower 96 Northeastern Naturalist Vol. 11, Special Issue 2 than one would expect for the amount of dissolved inorganic nitrogen in the water column. Cobscook Bay chlorophyll concentrations never exceeded 4μg/l (Phinney et al. 2004), a level considered oligotrophic to mesotrophic (National Research Council 2000, US Environmnetal Protection Agency 2001). Garside and Garside (2004) suggest that short residence times due to high flushing and heavy grazing by invertebrate communities may be preventing phytoplankton populations from fully developing. Enhancing phytoplankton populations is not the only potential manifestation of nutrient enrichment, however. Attached macroalgae and other benthic algae may be stimulated by nutrients when phytoplankton are not. Phinney et al. (2004) offer that nutrients may be utilized by macrophytes and benthic algae rather than phytoplankton. Vadas and Beal (1987) suggested that green algal mats in Cobscook Bay may be a result of nutrient enrichment from salmon pens. When their field work was conducted in 1984, however, salmon aquaculture production in Cobscook Bay was less than 5% of that of the mid-1990s. More recently, Vadas et al. (2004c) measured greater algae cover and biomass at one field site closer to a salmon pen and suggest a causal relationship. However, proximity alone is insufficient evidence to conclude a cause and effect relationship. Their sample location closest to the salmon farm was also located closest to nutrient sources outside Cobscook Bay (e.g., offshore and Passamaquoddy Bay). Also, algal growth was greatest in February, when salmon operations were nearly dormant and loading from aquaculture was minimal. Other information, empirical and anecdotal, further clouds the conclusion that macroalgae are being stimulated by human activities in Cobscook Bay. In the mid-1970s, Timson (1976) mapped and estimated that approximately 1062 hectares of flats were covered with green algae. Using different methods, Larsen et al. (2004) report an increase in acreage covered with green algae. However, over the years, we have interviewed many residents around Cobscook Bay who recall green algae mats that “come and go” and also that mats covered entire mudflats before the 1970s, well prior to marine salmon farming. Considering the different methods employed by the two studies and interannual variablity, we are uncertain if the difference is truly significant. Factors in addition to nutrient enrichment control algae (Lotze et al. 2001, Pihl et al. 1999, Trimmer et al., 2000, and Worm et al. 1999). Herbivory, abundance of overwintering propagules, ice scour, temperature, and physical disturbance also must be considered. We do not discount the possibility that green algae is increasing. However, based on the magnitude of nutrients available in Cobscook Bay from sources other than salmon farming, we believe this topic merits further study to determine if, in fact, green algae mats are nutrient limited. 2004 J.W. Sowles and L. Churchill 97 We assessed the third expression of eutrophication, hypoxia, from dissolved oxygen profiles collected as part of the annual environmental monitoring of finfish farms (FAMP 1988–2003) and a 1995 coastwide study of oxygen that included five locations in the Inner Bay (J.W. Sowles, unpubl. data). Inside and immediately adjacent (5 m) to the pens, small differences attributed to fish respiration are measured where oxygen, with few exceptions, is reduced by no more than 1 mg/l and 5% saturation less than what is measured at reference stations located 60 m upcurrent. Over 16 years of monitoring, oxygen concentrations 30 m beyond and beneath the pens have consistently remained above 7 mg/l and 85% saturation. On an ecosystem level, in the Inner Bay, the portion of Cobscook Bay we believed to be most sensitive to the effects of eutrophication, oxygen profiles taken at sunrise and mid-morning in August and September 1995 showed no signs of a diurnal cycle, and oxygen never fell below 7 mg/l and 85%. Both data sets lead us to conclude that hypoxia is not a concern in Cobscook Bay. Lastly, it is important to acknowledge that southwest New Brunswick’s salmon industry is located in Passamaquoddy Bay, adjacent to Cobscook Bay. There, the industry is presently and has historically been considerably larger than that inside Cobscook Bay (Department of Fisheries and Oceans 2003), suggesting that New Brunswick farms could be responsible for the large flux reported by Garside and Garside (2004). Concentrations of nitrogen and phosphorus measured at the entrance to Cobscook Bay by Phinney et al. (2004) are within the range of those measured offshore in Bay of Fundy surface waters (Bugden et al. 2001, Martin et al. 1999). Strain and Hargrave (2005) estimated the nitrogen flux from salmon farms in southwestern New Brunswick. They calculate that in some areas (e.g., Letang/Letete regions) change in nitrogen concentration due to salmon farms is large relative to the Fundy concentrations. However, they also predicted very small changes (0.2 μM nitrogen and 0.012 μM phosphorus) to ambient nutrient concentrations caused by farms closest to Cobscook Bay in Northern Passamaquoddy and off Deer and Campobello Islands. We conclude that the New Brunswick salmon farms probably had minor influence on our Cobscook Bay water quality prediction considering the magnitude of offshore sources. Conclusions Using two independent methods, we estimated that salmon aquaculture during the period 1995–1996 contributed an annual load of 360,000 kg nitrogen and 85,000 kg phosphorus to Cobscook Bay. However, although the nutrient load from salmon farms is large by absolute measure, Cobscook Bay’s position in a nutrient rich corner of the Gulf of Maine together with 98 Northeastern Naturalist Vol. 11, Special Issue 2 its extreme tides and currents appears to result in a negligible effect on planktonic primary productivity. As noted by others in this issue(Garside and Garside 2004, Phinney et al. 2004), water residence times are so short that phytoplankton do not accumulate even in Inner Cobscook Bay. Not only does the flux of water from outside Cobscook Bay dilute and flush nutrients and phytoplankton, unlike many other bays in the United States where local and landside nutrient sources drive eutrophication (National Research Council 2000), water beyond Cobscook Bay appears to be the dominant source of nutrients (Garside and Garside 2004). Conclusions regarding nutrient loading effects on intertidal macroalgae remain ambiguous, however. Further investigation on macroalgae within and beyond Cobscook Bay is important to the management and avoidance of adverse impacts from finfish aquaculture. Acknowledgments We thank Dave Brooks, Thomas Trott, Dan Canfield, and two anonymous reviewers who helped improve the paper by providing support and suggestions concerning our predictive approach. We are particularly grateful for the constructive comments and encouragement of Peter Larsen and Barbara Vickery. Literature Cited Ackefors, H., and M. Enell. 1990. Discharge of nutrients from Swedish fish farming to adjacent seas areas. Ambio 19:28–35. Ackefors, H., and M. Enell. 1994. The release of nutrients and organic matter from aquaculture systems in Nordic countries. Journal of Applied Ichthyology 10:225–241. Axler, R.P., C. Tikkanen, J. Henneck, J. Schuldt, and M.E. McDonald. 1997. Characteristics of effluent and sludge from two commercial rainbow trout farms in Minnesota. Progressive Fish-Culturist 59:161–172. Beveridge, M.C.M. 1996. Environmental impacts and environmental capacity. Pp. 152–186, In M.C.M. Beveridge (Ed.). Cage Aquaculture (2nd Edition). Fishing News Books. Oxford, UK. Brooks, D.A. 2004. Modeling tidal circulation and exchange in Cobscook Bay, Maine. Northeastern Naturalist 11(Special Issue 2):23–50. Brooks, D.A., M.W. Baca, and Y.-T. Lo. 1999. Tidal circulation and residence time in a macrotidal estuary: Cobscook Bay, Maine. Estuarine, Coastal, and Shelf Science. 49:647–665. Bugden, J.B.C., B.T. Hargrave, P.M. Strain, and A.R.J. Stewart (Eds.). 2001. Spatial patterns of some physical and chemical variables in Passamaquoddy Bay and Letang Inlet, southwestern Bay of Fundy. Spetember 1999. Canadian Technical Report of Fisheries and Aquatic Sciences 2356:iv + 96 pp. Campbell, D.E. 2004. Evaluation and emergy analysis of the Cobscook Bay ecosystem. Northeastern Naturalist 11(Special Issue 2):355–424. Cho, C.Y., and D.P. Bureau. 1997. Reduction of waste output from salmonid aquaculture through feeds and feeding. The Progressive Fish Culturist. 59:155–160. 2004 J.W. Sowles and L. Churchill 99 Department of Fisheries and Oceans, Science Branch, Maritimes Region. 2003. Salmon holding capacity in southwestern New Brunswick. Canadian Technical Report of Fisheries and Aquatic Sciences 2489: iv + 60pp. Enell, M., and H. Ackefors. 1992. Development of Nordic salmonid production in aquaculture and nutrient discharges into adjacent sea areas. Aquaculture Europe 16:6–11. Findlay, R.H., and L. Watling. 1994. Toward a process-level model to predict the effects of salmon net-pen aquaculture on the benthos. Pp. 47–77, In B.T. Hargrave (Ed.). Modelling Benthic Impacts of Organic Enrichment from Marine Aquaculture. Canadian Technical Report of Fisheries and Aquatic Sciences 1949:xi +125pp. Finfish Aquaculture Monitoring Program (FAMP). 1989–2003. Unpublished harvest and feed reports on file at the Maine Department of Marine Resources, West Boothbay Harbor, ME. Garside, C., and J. Garside. 2004. Nutrient sources and distributions in Cobscook Bay, Maine. Northeastern Naturalist 11(Special Issue 2):75–86. Garside, C., J. Garside, and P.F. Larsen. 2004. Nutrient sources and distribution in Cobscook Bay with data appendix. Bigelow Laboratory Technical Report Number 116. Bigelow Laboratory for Ocean Sciences. West Boothbay Harbor, ME. Gillibrand, P.A., M.J. Gubbins, C. Greathead, and I.M. Davies. 2002. Scottish executive locational guidelines for fish farming: Predicted levels of nutrient enhancement and benthic impact. Scottish Fisheries Research Report No. 63/2002. Fisheries Research Services, Marine Laboratory, Aberdeen, Scotland. Goldberg, R., and T. Triplett. 1997. Murky Waters: Environmental Effects of Aquaculture in the US. The Environmental Defense Fund. New York, NY. Gowen, R. 1994. Managing eutrophication associated with aquaculture development. Journal of Applied Icthyology 10:242–257. Gowen, R.J., and N.B. Bradbury. 1987. The ecological impact of salmonid farming in coastal waters: A review. Oceanography and Marine Biology: An Annual Review 25:563–75. Hardy, R. 1999. Problems and opportunities in fish feed formulation. Aquaculture Magazine 25:56–60. Larsen, P.F. 2004. Introduction to ecosystem modeling in Cobscook Bay, Maine: A boreal, macrotidal estuary. Northeastern Naturalist 11(Special Issue 2):1–12. Larsen, P.F., S. Barker, J. Wright, and C.B. Erickson. 2004. Use of cost effective remote sensing to map and measure marine intertidal habitats in support of ecosystem modeling efforts: Cobscook Bay, Maine. Northeastern Naturalist 11(Special Issue 2):225–242. Lotze, H., B. Worm, and U. Sommer. 2001. Strong bottom-up and top-down control of early life stages of macroalgae. Limnology and Oceanography 46:749–757. Martin, J.L., M.M. LeGresley, P.M. Strain, and P. Clement. 1999. Phytoplankton monitoring in the southwest Bay of Fundy during 1993–1996. Canadian Technical Report of Fisheries and Aquatic Sciences 2265:iv + 132 pp. National Research Council. 2000. Clean Coastal Waters: Understanding and Reducing the Effects of Nutrient Pollution. National Academy Press. Washington, DC. Naylor, R.L., R.J. Goldberg, J.H. Primavera, N. Kautsky, M.C.M. Beveridge, J. Clay, C. Folke, J. Lubchenco, H. Mooney, and M. Troell. 2000. Effect of aquaculture on world fish supplies. Nature 405:1017–1024. 100 Northeastern Naturalist Vol. 11, Special Issue 2 Phinney, D.A., C.S. Yentsch, and D.I. Phinney. 2004. Primary productivity of phytoplankton and subtidal microphytobenthos in Cobscook Bay, Maine. Northeastern Naturalist 11(Special Issue 2):101–122. Pihl, L., A. Svenson, P. Moksnes, and H. Wennhage. 1999. Distribution of green algal mats throughout shallow soft bottoms of the Swedish Skagerrak archipelago in relation to nutrient sources and wave exposure. Journal of Sea Research 41:281–294. Rottiers, D.V. 1993. Elemental composition of a migratory and a land-locked strain of Atlantic Salmon, Salmo salar. Comparative Biochemical Physiology 104a:93–100. Solbe, J. 1988. Water quality. Pp. 69–87, In Laird L. and T. Needham (Eds.). Salmon and Trout Farming. Ellis Horwood, Chichester, UK. Strain, P.M., and B.T. Hargrave. In press. Salmon aquaculture, nutrient fluxes, and ecosystem processes in southwestern New Brunswick. In Hargrave B.T. (Ed.). Environmental Effects of Marine Finfish Aquaculture. The Handbook of Environmental Chemistry Volume 5: Water Pollution. Springer Verlag. Berlin Heidelberg New York, NY. Timson, B.S. 1976. Coastal marine geologic environment maps. Maine Geological Survey Open File Report 77-1, Augusta, ME. Trimmer, M., D. Nedwell, D. Sivyer, and S. Malcolm. 2000. Seasonal organic mineralization and denitrification in intertidal sediments and their relationship to the abundance of Enteromorpha sp. and Ulva sp. Marine Ecology Progress Series 203:67–80. Trott, T.J. 2004. Late 20th-century qualitiative faunal changes in Cobscook Bay, Maine. Northeastern Naturalist 11(Special Issue 2):325–354. US Environmental Protection Agency. 2001. Nutrient Criteria Technical Guidance Manual: Estuarine and Coastal Marine Waters. EPA-822-B-01-003. Washington, DC. Vadas, R., and B. Beal. 1987. Green algal ropes: A novel estuarine phenomenon in the Gulf of Maine. Estuaries 10:171–176. Vadas, R.L., W.A. Wright, and B.F. Beal. 2004a. Biomass and productivity of intertidal rockweeds (Ascophyllum nodosum LeJolis) in Cobscook Bay, Maine. Northeastern Naturalist 11(Special Issue 2):123–142. Vadas, R.L., B.F. Beal, W.A. Wright, S. Nickl, and S. Emerson, 2004b. Growth and productivity of sublittoral fringe kelps (Laminaria longicruris) Bach. Pyl. in Cobscook Bay, Maine. Northeastern Naturalist 11(Special Issue 2):143–162. Vadas, R.L., B.F. Beal, W.A Wright, S. Emerson, and S. Nickl. 2004c. Biomass and productivity of red and green algae in Cobscook Bay, Maine. Northeastern Naturalist 11(Special Issue 2):163–198. Vinogradov, A.P. 1953. The Elemental Chemical Composition of Marine Organisms. Memorial Sears Foundation for Marine Research. No. 2. New Haven, CT. Wildish, D.J., P.D. Keizer, A.J. Wilson, and J.L. Martin. 1993. Seasonal changes of dissolved oxygen and plant nutrients in seawater near salmonid net pens in the mactrotidal Bay of Fundy. Canadian Journal of Fisheries and Aquatic Sciences 50:303–311. Worm, B., H. Lotze, C. Bostrom, R. Engkvist, V. Labanauskas, and U. Sommer. 1999. Marine diversity shift linked to interactions among grazers, nutrients, and propagule banks. Marine Ecology Progress Series 185:300–314. Young, K., F. O’Hara, C. Lawton, and C. Colgan. 1998. Report on current practices and benefits of finfish aquaculture in Maine. Prepared for the Department of Marine Resources. Hallowell, ME.