Final draft of a working paper prepared for the Global Programme of Action Coalition for the
Gulf of Maine (GPAC) and the Secretariat
of the Commission for Environmental Cooperation

This working paper was prepared by:

This working paper was prepared for the Global Programme of Action Coalition for the Gulf of Maine (GPAC) and the Secretariat of the Commission for Environmental Cooperation (CEC). The views contained herein do not necessarily reflect the views of the CEC, or the governments of Canada, Mexico or the United States of America.

Reproduction of this document in whole or in part and in any form for educational or non-profit purposes may be made without special permission from the CEC Secretariat, provided acknowledgment of the source is made. The CEC would appreciate receiving a copy of any publication or material that uses this document as a source.

© Commission for Environmental Cooperation, 1997

For more information about this or other publications from the CEC, contact:

Commission for Environmental Cooperation
393, rue St-Jacques Ouest, bureau 200
Montréal (Québec) Canada H2Y 1N9
Tel: (514) 350-4300 • Fax: (514) 350-4314
http://www.cec.org

Table 1. Matrix of potential impacts of land-based activities

on Gulf of Maine marine habitats 76

Table 2. Relative significance of impacts of land-based activities

on marine habitats based on selected attributes 77

Figure 1. Productivity of selected coastal habitats. Black portion

of bar indicates upper and lower values of range of available estimates 79

Figure 2. Frequency distributions (percent) of marine habitat concerns based

on interviews with community groups and CLF/CCNB compilations 80

Figure 3. Frequency distributions (percent) of land-based activity concerns based on interviews with community groups and CLF/CCNB compilations 81

Profile of the Commission for Environmental Cooperation

In North America, we share vital natural resources including air, oceans and rivers, mountains and forests. Together, these natural resources are the basis of a rich network of ecosystems which sustain our livelihoods and well-being. If they are to continue being a source of future life and prosperity, these resources must be protected. Protecting the North American environment is a responsibility shared by Canada, Mexico and the United States.

The Commission for Environmental Cooperation (CEC) is an international organization whose members include Canada, Mexico and the United States. The CEC was created under the North American Agreement on Environmental Cooperation (NAAEC) to address regional environmental concerns, help prevent potential trade and environmental conflicts and to promote the effective enforcement of environmental law. The Agreement complements the environmental provisions established in the North American Free Trade Agreement (NAFTA).

The CEC accomplishes its work through the combined efforts of its three principal components: the Council, the Secretariat and the Joint Public Advisory Committee (JPAC). The Council is the governing body of the CEC and is composed of the highest-level environmental authorities from each of the three countries. The Secretariat implements the annual work program and provides administrative, technical and operational support to the Council. The Joint Public Advisory Committee is composed of fifteen citizens, five from each of the three countries, and advises the Council on any matter within the scope of the agreement.

The CEC facilitates cooperation and public participation to foster conservation, protection and enhancement of the North American environment for the benefit of present and future generations, in the context of increasing economic, trade and social links among Canada, Mexico and the United States.

Vision

A healthy marine and coastal environment in the Gulf of Maine where human use and biological diversity thrive in harmony.

Mission

The GPAC will work with all interested parties to assist in the application of the Global Programme of Action for the Protection of the Marine Environment from Land Based Activities (GPA). This Coalition will draw from and build upon the existing work of the Gulf of Maine Council, the Regional Association for Research in the Gulf of Maine, the Commission for Environmental Cooperation (CEC) and other organizations and individuals committed to the protection of this shared public resource of world-class cultural, economic, ecological and intrinsic value.

The GPAC will assist public and private entities in the Gulf of Maine region identify pollution and habitat priorities and work to strengthen the capacity of these organizations and individuals to address them.

1998 Objectives

Identify and assess current knowledge on the marine and coastal habitats of the Gulf of Maine and the existing and potential effects of pollutants from land based activities on their sustainability.

Organize a workshop of multidisciplinary and cross-sectoral participants to review this knowledge and produce a consensus list of the priority pollutants and critical habitats in the Gulf of Maine requiring immediate action.

Identify strategies and measures related to the management of priority pollutants and critical habitats identified during this first workshop.

Organize a second workshop of multidisciplinary and cross-sectoral participants to assess management strategies and produce a regional response with immediate and long-term measures intended to reduce pollutants and protect and manage habitat in the Gulf of Maine. It will include financing mechanisms and a process for review and evaluation of implementation success.

Secure resources from interested stakeholders to begin implementation of actions to advance the elements of the Action Plan.

Results (late 1998-early 1999)

Broad-based, cross-sectoral stakeholder consensus on regional habitat and pollutant priorities and commitment to responding to them.

Implementation begins, within and across jurisdictions, including select demonstration projects.

Transitional seed financial support from the CEC for implementation.

Strengthened binational commitment to GPA implementation.

Conclusion of GPAC role as regional stakeholders initiate implementation.

1. Introduction

Some 3.6 million people dwell in the coastal regions rimming the Gulf of Maine (Dow and Braasch 1996). It is safe to say that all derive some degree of benefit from living near a healthy productive marine ecosystem. The livelihood of many depends on the diverse and valuable harvests of finfish, shellfish and marine plants. In recent years fisheries in the Gulf have directly involved some 20,000 individuals in harvesting over 500,000 metric tons of fish and shellfish valued at $650 million (US) each year (Apollonio and Mann 1995). The indirect economic impacts of the fisheries are many times this. In recent decades the farming of fish and shellfish has become a major industry in many coastal areas. Aquaculture is an industry whose existence depends on the quality and productivity of nearshore waters. The shorelines and waters of the Gulf also provide a wide range of recreational opportunities for most residents, and each year several million tourists flock to the region to share many of the same seashore recreational pursuits. The Gulf is also an important corridor for commercial shipping, as evidenced by the many busy ports scattered around its perimeter. With due respect for their finite assimilative capacity, the productive and endlessly circulating waters of the Gulf can also continue to safely disperse and degrade some of the wastes generated by human populations. There seems little doubt that our incentives for sustaining a healthy and productive ecosystem in the Gulf of Maine should be many and compelling.

However, there are indications that the health and productivity of this valued marine system are being increasingly compromised by a wide range of human activities occurring along its coasts and throughout the Gulf's immense watershed. These stresses on the ecosystem are not all of recent origin; many have been steadily building in the centuries since the first European occupation of the region in the early 1600s. The early aboriginal inhabitants were relatively sparse, used simple technologies and thus had minimal impacts on coastal ecosystems. In contrast, Europeans, almost from the outset, began altering the natural landscape on a massive scale. As populations expanded and technologies advanced during the ensuing centuries the rate of alteration of the landscape increased rapidly (Gordon 1989). Although there is little quantitative information about the early state of the Gulf's coastal ecosystem, historical accounts unequivocally show that many coastal fish and wildlife populations have since been decimated and large tracts of once productive marine habitats lost or degraded. Even so, many of these ecosystem perturbations in the Gulf are still relatively minor in comparison to the severe impacts evident in many other coastal regions around the world. Significant areas of the Gulf can even truthfully be characterized as still "relatively pristine and healthy" (Dow and Braasch 1996).

The most pressing problems are largely confined to coastal waters, particularly in estuaries and harbors (Dow and Braasch 1996), and in areas where populations densities are high or rapidly expanding. However, if the environmental health of significant coastal ecosystems continues to deteriorate, it is likely that in time the ecological integrity and productivity of much of the Gulf may be compromised. Concerted efforts are needed to stem the alarming rate of habitat loss and, where feasible, to restore some of the critical ecological functions to degraded habitats. Many of the undesirable habitat changes that are occurring are clearly a direct consequence of land-based human activities, not only in coastal areas but throughout the watershed. It may be possible to reduce or reverse the adverse impacts of some of these activities on coastal habitats if we act promptly to better understand the causes and effects and take decisive steps as a society to better manage our social and economic affairs.

The goal of this scoping paper is to present an overview of the land-based activities associated with human occupation of the Gulf of Maine area that may be contributing to loss or physical degradation of important marine habitats. The pervasive effects of contaminants are dealt with at length in a companion scoping paper. As well as reviewing the available scientific literature concerning the environmental issues, the paper also considers the views of knowledgeable residents and groups in the region about habitat changes and the land-based activities that may be responsible. To be successful, any efforts to resolve the environmental problems besetting the Gulf must involve, and have the support of, citizens and their communities and must address their particular concerns and aspirations. The report concludes by identifying several important environmental issues, tentatively ranking them and making some preliminary recommendations for appropriate management objectives and actions for dealing with them.

The Global Program of Action for the Protection of the Marine Environment from Land-Based Activities (Global Program of Action or GPA), was adopted by the United Nations Environment Program on November 3, 1995. The GPA aims to prevent the degradation of the marine environment from land-based activities by promoting and facilitating the preservation and protection of the marine environment by States. It is designed to assist States in taking actions, individually or jointly within their respective policies, priorities and resources, which will lead to the prevention, reduction, control and/or cessation of the degradation of the marine environment, as well as to its recovery from the impacts of land-based activities. Achievement of the aims of the Program of Action will contribute to maintaining and, where appropriate, restoring the productive capacity and biodiversity of the marine environment, ensuring the protection of human health, as well as promoting the conservation and sustainable use of marine living resources.

The GPA calls for actions by each signatory nation to preserve and protect the marine environment on a national, regional and international basis in order to reach the goal of "sustainable seas". In North America, the Commission for Environmental Cooperation (CEC) was created as a result of NAFTA (North American Free Trade Agreement) negotiations to facilitate cooperation and public participation and to foster conservation, protection and enhancement of the North American environment.

In pursuing its mandate, the CEC decided to promote a series of pilot projects in North America to implement the GPA, and selected the Gulf of Maine (GOM) as a candidate site for one of the projects. CEC brought together an diverse group of individuals with an interest in the GOM and the GPA to develop and implement a project of their own design, with some support from the CEC. The group, which has named itself the GPA Coalition on the Gulf of Maine, has formulated an action plan to this end. A key component in the plan is a workshop in Saint John, New Brunswick on April 29 and 30, 1998 at which participants will focus on prioritizing habitat issues in the Gulf of Maine. The participants will include industry, community groups, municipalities, scientific institutions, and provincial, state and federal governments. This scoping paper has been prepared to assist participants at that workshop by providing background information on land-based activities and related habitat issues and by recommending possible management objectives and actions for consideration.

The intent of this scoping paper is to provide an overview of the information available about land-based activities and their potential effects on marine habitats in the Gulf of Maine. The land-based activities considered are defined and described in section 4, while the habitats and habitat issues are dealt with in section 5. Based on an assessment of the scientific information, and on a consideration of community concerns, several priority issues are identified in section 6 and a number of possible management objectives and actions are outlined in section 7. A number of factors often obscure the causal links between land-based activities and observed habitat degradation and loss. These factors include confounding natural processes, such as ecosystem variability involving unpredictable, cyclical or progressive changes, multiple sources of stress involving cumulative or synergistic effects, as well as a suite of complex and poorly known intermediary processes such as eutrophication , sediment dynamics, hydrology and habitat regeneration. Some of these are briefly described in section 4.7 in order to explain the overall approach used in this scoping paper .

Because of this difficulty in obtaining unequivocal evidence to link most land-based activities with specific changes in habitats, this report adopts a two-pronged approach. First it looks at various land-based activities and the possible environmental stresses generated that could perturb marine habitats (section 4) . Then it looks at the principal habitat types (section 5) and attempts to identify the stresses that might potentially be associated with particular land-based activities. Finally, by assessing the situation from both these perspectives, it uses an activity-habitat matrix to provide aN assessment of the probable significance of ecological impact linkages between specific land-based activities and particular marine habitats. This involved A review OF the voluminous scientific literature available and consultation with specialists in many fields.

In addition to an examination of the scientific data, two other sources of information about the potential impacts of land-based activities on marine habitats in the Gulf of Maine were considered. To gauge general community perceptions of what habitats are being lost, degraded or at risk in their local area, and which land-based activities are considered to be environmentally threatening, we interviewed knowledgeable representatives from 48 different groups distributed widely around the Gulf. These included conservation, environmental, First Nations, fisher/harvester, industry and research groups (Appendix 1). A standard questionnaire was used for the phone interviews (Appendix 1). Complementary information on habitat degradation and land-based activities was also extracted from detailed compilations of information on all major Gulf of Maine Estuaries being prepared by the Conservation Law Foundation (Shelley in press) and the Conservation Council of New Brunswick (Harvey in press) for the Gulf of Maine Estuaries Restoration Project component of the Restore America's Estuaries Program. The detailed information on each estuary had been compiled with the assistance of individuals particularly knowledgeable about the local area and its environmental issues. The information obtained from the interviews and the estuary compilations was first analyzed separately and then combined to provide an overview and relative ranking of community concerns regarding habitats at risk and human activities responsible. These preliminary rankings of community concerns were then carefully considered in conjunction with the available scientific information about the probable significance and extent of linkages between causes and habitat effects and by this means a series of priority issues for the Gulf region was identified and ranked. The ranking was based on a somewhat subjective assessment of the combined socioeconomic and ecological significance of the issues identified.

Human activities immediately adjacent to the coastal zone typically have the most devastating and readily observable impacts on estuarine, intertidal and inshore habitats. However, it is increasingly being recognized that activities throughout the 179,000 km² of land area that drains into the Gulf (Dow and Braasch 1996) can significantly effect coastal habitats, often in subtle and poorly understood ways. The waterborne transport of industrial contaminants and other noxious materials is particularly worrisome in this regard, and is treated at length in the contaminant scoping paper. In addition, there are a variety of physical alterations in habitats, involving either degradation, actual loss or functional loss that can also be attributed to human activities in the watershed. It is these physical effects that will be considered here. The land-based activities discussed are largely those identified in the Global Plan of Action (UNEP 1995). However, an activity not specifically identified in the GPA; namely heavy mobile fishing gear use in inshore waters, is also considered because of growing scientific and public concern about the threat posed to inshore benthic habitats by what is arguably a land-based activity.

The ‘culture’ of human activities includes both intentional actions that involve decision-making as well as largely unintentional ones such as growth in population. Growth in population has a fundamental influence on the health of coastal habitats. Perhaps more significantly, growth in another dynamic process; namely urbanization, often far exceeds the numerical growth in population. Population growth and urbanization are distinct and synergistic processes calling for different response strategies. There will inevitably be population growth and increasing urbanization in the Gulf of Maine region. Whether this involves growth in population, no growth but changes in land use, or a relative population shift to the coastal zone, increasing human activities and changes in patterns of land use will have steadily growing impacts on coastal habitats. A pervasive concern associated with coastal development is the cumulative impacts not only on the environment but on the economy and community identity. Though this paper focuses primarily on habitat impacts, it is important to keep in mind these broader economic and social implications of habitat loss and degradation.

The coast has traditionally been the area of concentration of population worldwide, and is also the area of the United States where growth, three times the national average, is at its highest (Colliton et al. 1992; LMER 1992). The population density within 50 miles of the Gulf of Maine varies greatly in different regions, from 0.1 to 10,000 people per square kilometer (US Census Bureau 1996), with Massachusetts coastal counties accounting for two thirds of the total population (Colgan 1989). This differential is expected to continue to increase, with the three US states growing at four times the rate of New Brunswick and Nova Scotia (Colgan 1989). Projections suggest that the New Hampshire population will increase by 25% by the year 2025, while those of Massachusetts and Maine will rise by 15% (ACBJ Research 1997). Growth rates will be highest along the coast as far north as Lincoln County, Maine, with much lower, stable or even negative growth rates in Northern Maine and the Canadian provinces (Colgan 1989).

The coastal populations are increasingly becoming concentrated in a few major centers and, perhaps not surprisingly, environmental problems are most acute in these same areas (Dow and Braasch 1996). The major centers of population grouped in four descending size categories are 1) Boston; 2) St. John and Portland; 3) Moncton, Bangor, Lewiston, New Bedford and Fall River; and 4) Truro, Waterville, Augusta, Bath, Brunswick, Biddeford, Dover, Portsmouth, Haverhill, Lawrence, Gloucester, Salem, Lynn, Cambridge, and Quincy (Rand McNally 1993). Boston is by far the largest urban center on the Gulf coast and, together with St. John and Portland, is significantly greater in population density so as to have very different impacts on coastal habitat than the smaller centers. However, Boston and St. John are presently experiencing low growth rates or even declines in population. which could eventually affect their financial ability to adequately address ecological problems (Colgan and Plumstead 1995).

Urbanization is not limited to the larger urban and suburban centers, but is spreading through many rural counties. Projections indicate that two of the three largest urban centers will not be absorbing a proportionate amount of the anticipated rise in the population of Gulf region, and that increases in counties immediately north of Boston will not exceed the state average. In Nova Scotia, due to its compact geography, Halifax, located on the Atlantic coast, is catalyzing growth in three counties bordering the Bay of Fundy (Colgan and Plumstead 1995). A trend in both countries that could have a major impact in the Gulf region is the marked increase of micropolises , or centers of 10,000 to 50,000 population (ACBJ Research 1997). This can be anticipated throughout the region given the abundance of attractive small towns along the coast. Another trend is a rapid increase in the proportion of housing starts in rural counties. In the Gulf of Maine area, this is significant in at least one county and possibly in others. On Cape Cod housing starts have increased at a rate of 36% compared to the state average of 10%, with Boston itself slightly below the state average (Valiela et al. 1992). These trends are fueled by rising incomes in urban centers, where the per capita income increases are the highest in the Gulf region (Colgan and Plumstead 1995). In the urbanizing rural counties, growth in housing starts is parallel to economic growth in the trades and services sector, which takes up somewhat more land than an equivalent economic growth in the manufacturing sector (Colgan 1989). This multiplies the effect of population growth substantially. Based on studies of rural New York and Connecticut counties, it has been shown that urban land use expanded at a rate up to eight times greater than the population growth. With urban land patches fragmenting the landscape at an increasing rate, the EPA has now identified landscape fragmentation as one of its top priorities (LaGro 1994; Chester Arnold pers. comm.). In many regions of the Gulf, rural counties will increasingly be called upon to deal with pressure of increasing urbanization and landscape fragmentation with an infrastructure and bureaucracy ill-equipped to handle it.

Land use planning responses to coastal development in different parts of the Gulf of Maine is varied, ranging from few, mostly ineffective, controls to very stringent ones that are strictly enforced. However, there is a growing recognition of the need for more effective protection of the coast and for better control of growth and development. There is rising interest in the development and use of innovative approaches to the problems. For example, the Massachusetts Bays Program has created a water quality modeling tool called FecaLOAD to estimate the fecal coliform loading from storm water runoff. It has been successfully applied in Casco Bay. New Brunswick proposed a provincial coastal policy that addresses development and use of the coast and inshore waters. Brunswick ME enacted a Coastal Protection Zone applicable to fragile coastal embayments. A 1988 study had shown that a severe shellfish kill was caused by nutrient loading from municipal waste water, septic fields, fertilizers and stormwater runoff. The new zoning proscribes setbacks, larger lots, improved septic design and maintenance and fertilizer use (Vestal and Rieser 1995). There are similar examples from throughout the region, suggesting that the desire for positive change is strengthening.

A poll of 333 coastal zone specialists at the Coastal Zone ’97 Conference in Boston revealed that concern for the issue of existing and emerging ‘megacities’ outweighed that for climate change by 73% (Intercoast Network 1998). Likewise, the issue of development, population growth, and human activity in the Gulf of Maine coastal zone is perceived as one of the greatest threats to the health and degradation of the coastal habitats and their associated economies. Human land use activities and urbanization causes increased habitat fragmentation, degradation and loss in the coastal zone by:

- replacement of natural habitats with buildings, roads, recreational facilities and artificial landscaping

- the introduction of ‘domestic’ flora and fauna into natural habitats

- diminished productivity; loss of coastal agriculture, forest and wetlands

- non point source pollution carried by storm water runoff

- protective measures such as construction breakwaters, dams and barrages and draining, diking or infilling of flood plain areas etc.

These latter are especially prevalent near urban centers where land (for commercial, industrial and residential needs) and land prices are at a premium (C. Drysdale. pers. comm.). Public access sites, recreational use (see section 4.6) and residential development is increasing along the shoreline and on coastal islands (see section 5.7). Developers recognize the value of living on the coast and consequently perpetuate ‘sprawl’ development, raising many issues related to private and public property rights. States and provinces have taken steps to ensure public access to the shore by establishing and maintaining park sites, access sites, and planning for future parks and access (written into many local and regional land use plans and strategies). In some areas it has included shorefront acquisition and improvement projects such as expanded or upgraded parking facilities, boat docking facilities and beach access.

Many studies have indicated that another major effect of urbanization is the dramatic reduction in the extent of permeable landscape (Yamaguchi and Tyrell 1997). Streets, parking lots and roofs make up a substantial proportion of the new human landscape. It has been estimated that runoff doubles when natural ground cover is replaced by 10-20% impermeable surface (Intercoast Network 1998). In the Tidelands region of Connecticut, research for the Chester Creek Watershed Project showed that there is a strong correlation between the amount of impervious surfaces (concrete, asphalt, rooftops) in a watershed and the health of a receiving stream. Even in a watershed with a largely rural character, large-lot residential zoning, and extensive open space, the potential exists for significant degradation of water resources (Nelson and Arnold 1996).

Other concerns that are associated with expanding coastal development include:

- ecosystem and site specific health implications such as bacterial contamination from sewers, overboard discharge and urban stormwater runoff which results in periodic closure of shellfish areas to harvesting and beaches to swimming

- use and diversion of fresh water from coastal areas and its impact on the quality and quantity of potable water. This is of great concern with the increased development on coastal islands

- port and harbor development is increasing and major centers are developing strategies ranging from site specific (small bays and marinas) to comprehensive plans that consider adjacent areas of land and water that are impacted by the port activities and development (Boston Harbor). In most areas the principal focus is on ‘human use’ and less on ecosystem concerns (see section 4.2 regarding port dredging).

A disturbing trend in the last two decades has been an increase in shellfish bed closures near suburban and rural areas (NOAA 1986; 1991). For example, in Maine many of the 255,608 acres of shellfish beds are periodically closed to harvesting, and other coastal areas are often closed to swimming, because of bacterial contamination (Maine State Planning Office 1997). Urbanized land use is also a significant contributor to non point source pollution and nutrient loading of the coastal waters. In Waquoit Bay, domestic waste water from septic tanks provides more nitrogen than precipitation or use of fertilizers. This is compounded by the deforestation attendant with urbanization (Valiela et al. 1992; Valiela and Costa 1988; Persky 1986).

Urban centers also generate a great deal of garbage and other solid wastes and inevitably some proportion of it finds its way into the marine environment. Marine debris is pervasive throughout the Gulf of Maine. Marine debris comprises " any man made substance that enters the marine environment that does not readily biodegrade" (EPA website <http://www.epa.gov/owow/estuaries/coastlines/coastlines6.3/ marine.html>). It is not know how much marine debris there is in the Gulf as a whole, because an unknown proportion of it is "at sea", either on the bottom or floating offshore. The only real indication of the scope of problem is the substantial quantity that continually washes ashore. Volunteer beach clean up programs, of which there are many around the Gulf, not only remove much debris from shorelines, but also provide valuable information about the nature of the debris. This is important for identifying general sources and in designing effective strategies to reduce inputs. Typical beach sweeps shows that the debris comprises plastic (59%); paper (12%), glass (12%), metal (11%), wood (3%), rubber (2%), cloth (1%) (Centre for Marine Conservation Web Site: <http://cmc-ocean.org/mdio/sources.html>). Plastic, glass, metal and rubber are considered persistent debris because they are not readily degraded. The surveys suggest that 70-80% of the material comes from land-based sources, with much of it apparently washed through storm drains, trash from beach goers and material dumped illegally. The remainder comes from marine sources, typically, commercial fisheries, recreational boating and large vessel operations, particularly cargo ships.

Plastics are of great concern, because of the volume involved and the nature of the material. The features that make them so useful and widely used; namely their lightness, versatility, low cost, durability and strength, are the very things that make them a problem in the marine environment. Much of the plastic found is some form of packaging material, (bottles bags, lids etc.) or fishing gear (nets, rope, floats, pails etc.). The discarded nets, ropes and traps as well as cargo vessel wastes such as plastic strapping bands are particularly worrisome because they entangle and kill fish, turtles, birds and mammals. One beach survey also revealed that entanglement in fishing line discarded by recreational fishermen accounted for one third of all entangled animals found (CMC Web site). Some plastic debris (especially plastic bags and small fragments of Styrofoam) is often mistaken for food by marine animals and ingested, causing varying degrees of distress or death. In addition to being an aesthetic problem, debris on beaches also interferes with their recreational use by humans, sometimes at considerable cost. It is estimated that beach closures due to the presence of garbage and biomedical wastes resulted in the loss of over a billion dollars in tourism revenue in New York and New Jersey in 1987-88 (EPA Web site).

Controlling marine debris is difficult. Most jurisdictions in the region have regulations to control waste disposal and shoreline dumping, and treaties such as the International Convention for the Prevention of Pollution from Ships (MARPOL 73/78 Annex V) regulate the disposal of garbage at sea. However, the laws are difficult to enforce and it is virtually impossible to identify the specific sources of most debris, as it can travel great distances with currents. Public education, local litter control programs and volunteer clean up initiatives, probably offer the best hope for improving the situation in the long term. The Gulf of Maine Council on the Marine Environment has been in the forefront of many such efforts in the region (GOMCME 1996).

When Europeans first settled around the Bay of Fundy in the early 1600s there were an estimated 395 km2 of tidal salt marsh, largely fringing the upper reaches of the Bay (Gordon and Cranford 1994). They were quick to recognize the agricultural potential of the rich marsh soils and were also well acquainted with European marsh reclamation techniques. Over the next two centuries, the greater part of these marshlands were diked with earthen barriers, drained and tilled for crop production. The history of these monumental reclamation efforts up to recent times is concisely told in a publication by the Nova Scotia Department of Agriculture and Marketing (NSDAM 1987). During the 1940s a major dike restoration program was begun in the Maritimes, and over the next 20 years some 373 kilometers of dikes, protecting over 33,000 hectares of farmland were repaired or constructed. During the 1950s and 60s tidal barrages were constructed on most of the rivers around the Bay. These served to protect upstream reclaimed marshland from tidal flooding and reduced the need for frequent, costly repairs to miles of protective dikes. Today only about 65 km2 of tidally flooded salt marsh remains (Gordon and Cranford 1994), a mere 16% of the original area. Although New England salt marshes were diked to a lesser degree, large areas have, nevertheless, succumbed to other reclamation activities such as infilling, draining and ditching (Nixon, 1982). In many areas, dikelands are now being used for industrial and residential purposes, a worrisome trend in view of their vulnerability to storm surges and sea level rise (Parks et al. in press).

Given the paucity of quantitative information about early ecological conditions in the Fundy region it is difficult to assess the impacts of these massive diking projects on the marine ecosystem. Undoubtedly they influenced sediment dynamics and benthic habitats over large areas. Many studies (Gordon and Cranford 1994; Gordon et al, 1985; Adam 1990) have demonstrated the important role that salt marshes play in releasing organic detritus, and thus in enhancing the secondary productivity of the adjacent marine ecosystem. We shall probably never know how much more productive these coastal ecosystems were when the marshes were as much as six times more extensive, but undoubtedly the enhanced agricultural production has been achieved at the expense of substantial marine production.

For centuries, throughout the Gulf region, many high salt marshes that escaped diking have been heavily used as pasture for livestock or harvested as salt hay for fodder. Many of the early settlement patterns in New England were largely dictated by proximity to such hay marshes (Nixon 1982). In order to facilitate the harvest, many of these productive marshes were crisscrossed with ditches several decades ago to drain the land. In other marshes, particularly those near heavily populated areas, even deeper ditches were excavated in an effort to remove standing water to control mosquito populations. Such ditching projects were particularly prevalent during the depression as make work programs, so that by the late 1930s about 90% of the marshes of the northeast US coast had been extensively ditched (Nixon 1982). Such excessive drainage severely impaired the natural functioning of the salt marshes. Some of the adverse habitat impacts of grazing, harvesting and ditching are described in Nixon (1982) and Adam (1990). Now, with little market for salt hay and improved methods of mosquito control, there is growing interest in blocking the ditches in some marshes in an effort to restore some of the ecological functions.

A century or more of infilling of coastal wetlands with rock, soil and other material has already resulted in the irretrievable loss of large areas of many types of coastal habitat, particularly in the more populated areas of the Gulf (Platt et al. 1995a). Dredge spoils were often pumped directly onto salt marshes to create new land for industrial or residential development. For example, in Wells Harbor ME, 90 acres of marsh were filled with sediments dredged from the river during creation of a harbor (Van Dusen and Johnson Hayden 1989). Eelgrass beds and benthic habitats are also vulnerable to such infilling (see sections 5.8 and 5.10). In recent decades there has been a growing appreciation of the ecological value of such wetlands, and in most jurisdictions legislation now largely protects such areas from large-scale direct encroachments (Dow and Braasch 1996). However, there are indications that small-scale encroachments still occur in many places, both legally and illegally, as private landowners seek to expand or protect valuable and vulnerable coastal properties. Such small, incremental losses of marine habitat, seemingly innocuous in themselves can have substantial cumulative impacts over time.

Dredging, widespread in many parts of the Gulf of Maine, involves removing large volumes of sediment, often to a depth of several feet, from the ocean floor and depositing it elsewhere, either on land or in coastal waters. Two methods of dredging are currently widely used. Mechanical dredging employs large scoops or conveyers to physically scoop up sediment and place it in a barge for transport to a dump site. Hydraulic dredging involves loosening the sediment with water jets or rotating cutter heads to create a slurry, which is then pumped into a barge or directly through a pipeline to a disposal site. This latter is the most efficient, economical and commonly used (Kennish 1992). Dredging is routinely carried out in major port areas to maintain adequate draft for commercial vessels. Thus Boston, Portland, Portsmouth and St. John harbors have been regularly dredged over many decades. At present, a proposal for another major dredging program in Portland Harbor is being considered (Shelley in Press). Other navigable waterways such as the Cape Cod Canal, Salem Sound and the Kennebec River as far as Augusta, as well as many smaller ports, are also periodically dredged to remove accumulated sediments. Since 1968 almost a million cubic yards of subtidal sediments has been dredged from the Kennebec estuary and it is anticipated that at least another 600,000 cubic yards will be removed over the next 25 years (Shelley in press). Even areas remote from large ports may be subject to regular dredging activity in an effort to enhance recreational boating and support the development or expansion of marinas. Thus in 1974, over 11 acres of subtidal habitat in Bucks Harbor, Machias Bay was dredged to enhance access for recreational boating (Shelley in press). Similar dredging has taken place in parts of Casco Bay as well as in many of the other tourism "hotspots" along the coast.

Dredging may adversely effect estuarine and coastal habitats in a number of ways. These include primary, direct effects on benthic habitats, secondary effects on water quality and a variety of poorly understood indirect tertiary effect (Kennish 1992). The most devastating effects are clearly on benthic habitats, where the habitat and its biological communities are removed and destroyed, and on the terrestrial or submarine habitats that may be smothered by the dumped dredge spoils. The rate of recovery of the sediment structure and biological communities in the affected habitat varies with the nature of the habitat and the extent of the disturbance. Recovery of the benthic habitat from a major perturbation usually involves a series of successional stages (Rhodes et al. 1978), and typically, the same species assemblage returns to the area over time (Kennish 1992), unless the sediment composition has been drastically altered. Estimates of recovery times of animal communities in various places range from 6 months to over 3 years (Kennish 1992). Frequent dredging of an area may prevent the habitat from recovering to a stable, long-term equilibrium state.

A variety of secondary effects can adversely impact nearby habitats. Plant communities, such as eelgrass beds, not directly destroyed by dredging may be extensively damaged by slumping of sediments near the dredge site(Fred Short pers. comm.) and growth and survival over even more extensive areas can be impaired by turbidity and sediment deposition (Kennish 1992). Excessive fine sediment deposition can also smother shellfish beds and other subtidal and intertidal habitats. The dredging activity may also resuspend long-buried organic matter, nutrients and contaminants causing increased turbidity, eutrophication, increased biological-oxygen-demand as well as toxic effects on sensitive species (Engler 1990). Even more indirect and unpredictable tertiary effects on currents velocity and direction and on sediment transport may occur as a result of significant alterations of bottom topography as a result of dredging (Dow and Braasch 1996). This could in turn alter nearby benthic habitats or enhance erosion of sand beaches or estuarine shoreline.

Against these many negative impacts of dredging must be weighed a few possibly beneficial ecological effects. These include enhancing water circulation in estuaries, increasing the concentration of nutrients in the water and removing contaminated sediments from benthic habitats(Kennish 1992). For example, in Blue Bay in Maine the bottom sediments contain large amounts of sawmill wastes and contaminants that are undoubtedly adversely effecting the inshore benthic community (Shelley in press), and removal of these wastes by dredging might enhance the habitat in the long run. Dredging has also served to provide a ready supply of inexpensive material for use in landfilling or beach nourishment. However, the manner of disposal of such dredge "spoils" is a contentious issue, as existing habitat is inevitably smothered and destroyed. In fact, spoil disposal issues are becoming of even greater concern that the actual dredging itself. Contaminated sediments are nowadays virtually all deposited on terrestrial sites (Kennish 1992), but rarely, if ever, on coastal marshes and wetlands. Clean sediment is often disposed of in selected open water locations where it either remains (retentive sites) or is washed away by currents (dispersive sites). Some of the extensive research that has been conducted by the US Army Corps of Engineers on the impacts of dredging and spoil disposal on coastal habitats is summarized in Kennish (1992).

Humans have extensively modified coastlines in many part of the Gulf by constructing a variety of coastal structures for purposes ranging from shoreline protection to enhancement of shipping access. These structures include groins, breakwaters and jetties or wharves. Groins are barriers, set at right angles to the shore, usually extending from the backshore well into the subtidal zone (Lowenstein 1985). They may be constructed of timber, steel, concrete and/or quarry stone. These structures are designed to retard sediment loss or to trap sediments drifting along the shore, preventing erosion of an existing beach or creating a new one. Sometimes a series of them are constructed, forming a groin field or system , along a length of shoreline needing protection. Although they are usually effective in fulfilling their primary function of stabilizing the immediately adjacent beach, they more often than not exacerbate the erosion of beaches further down-current along the shore (Bird 1983).

Jetties are usually constructed near the mouths of inlets, bays or rivers to stabilize the navigation channel by controlling sediment movements, shield vessels from waves and provide a convenient place for docking vessels. Like groins, they may be constructed of timber, steel, concrete and/or quarry stone, and may in a similar fashion enhance erosion of downstream beaches or flats. Breakwaters are similar barriers, primarily designed to protect land or water areas behind them from the direct influence of waves. They may be constructed from a variety of readily available materials ranging from sunken ships to large concrete-filled fabric bags. There are two main types of breakwaters. Shore-connected ones protect a shore area, harbor, anchorage or basin from wave attack. Offshore ones are usually oriented parallel to the shore to provide wave protection to the shoreline in the lee of the structure.

Many thousands of these types of structures have been constructed in bays, inlets and estuaries along the length of the Gulf of Maine coast over the past two centuries or so. Many have virtually disappeared over the years, their resistant skeletons often a bleak reminder of changing economic circumstances. Large numbers of others have been all but abandoned and are poorly maintained and in various states of decay. They have all had, and most continue to have, an influence on the currents and the sediment dynamics in their immediate vicinity and thus on the structure of nearby subtidal and intertidal habitats (Mulvihill et al. 1980). The cumulative effects of such structures on marine habitats, especially when considered in relation to the many other structures usually found nearby that impede currents or restrict tidal incursions, may be significant. However, the ecological effects of such small-scale structures have not been well studied.

In many areas around the Gulf, where coastal erosion or storm wave attack threatens valuable real estate, various forms of protective structures have been constructed along the shoreline. Bulkheads, revetments and seawalls are all soil or sand retaining structures, designed to resist wave attack and are variously constructed of steel, wood, concrete, gabion baskets, quarry stone or rubble. Such structures are often built adjacent to vulnerable beaches and dune systems as described in section 5.4, where they often cause unforeseen problems (Lowenstein 1985). The total extent of coastline affected by various types of armoring is difficult to assess, as are the impacts of the structures on nearby intertidal and subtidal habitats. Most such structures have been in place for many decades or more, and marine habitats in their vicinity have undoubtedly long since attained a new equilibrium. There is a growing recognition that in many instances such protective armoring may be ecologically inappropriate. Maine, for example, now bans the construction of new seawalls on sand beaches, although existing ones can be maintained (Lowenstein, 1985; Joseph Kelley pers. comm.).

In many parts of the southern Gulf, wide sand beaches are critically important to the local tourism industry. Where such beaches are being rapidly eroded by waves and currents a program of beach nourishment or replenishment is often initiated. This involves importing large volumes of sand from terrestrial or marine (dredge spoils) sources to replace that being lost. In most situations such replenishment is little more than a quick fix, as the erosional forces are still present, and sometimes the imported sand erodes away even quicker than the native variety, leading to a ceaseless round of replenishments (Lowenstein 1985). The potential ecological impacts of such replenishment of sand beaches are explored more fully in section 5.4.

Dams are barriers, generally of rock or concrete, built across watercourses primarily to impound freshwater in upstream headponds for purposes of power generation, agricultural irrigation or downstream flood control. Most are fitted with mechanisms for releasing water in a controlled fashion. Over the past two centuries dams of varying sizes and types have regulated the patterns of flows of almost every significant river entering the Gulf of Maine. Early on, many dams were built to provide mechanical power for the thousands of sawmills that sprang up to supply lumber for the booming shipbuilding industry (Platt et al. 1995a). Many other industries also depended heavily on water power, so it is not surprising that population centers developed and expanded rapidly on the rivers where most such industries tended to concentrate. Later, dams were primarily built for the generation of electricity and to regulate downstream water levels to prevent flooding. Inevitably, their greatest proliferation has been in areas of greatest population densities and industrial activity; namely, the more southerly regions of the Gulf (Gordon 1989). Many rivers today have several dams along their length. For example the Saco River in Maine has 11 dams along a 70 mile stretch, while the Union River in eastern Maine reportedly has 42 dams along it’s length (Shelley in press).

The damming of rivers inevitably has important ecological repercussions, both upstream and downstream of the structure, largely as a result of interference with the transport of both water and waterborne sediments. By reducing the volume and rate of freshwater flow, and also by altering the patterns of flow at different times of the year, dams can significantly alter the hydrologic regimes in estuaries far downstream as well as in adjacent coastal waters. (Carter 1988). The changed flows can influence mixing and circulation patterns, salinity distributions, ambient water temperatures, ice formation and breakup and also nutrient concentrations (Fefer et al. 1980; Carter 1988; Platt et al. 1995a). The precise nature of these effects at specific dam sites are poorly understood because of the complexity of the dynamic processes involved.

Dams also impede the natural movements of sediments in the river, both in terms of bed transport (along the substrate) and suspended sediment transport (Carter 1988). Headponds often serve as a sediment trap and thus the dams interrupt the normal sediment supply to estuaries and nearby coastal areas. As a result, estuarine and coastal habitats that rely on a dynamic balance between sediment supply and loss, such as salt marshes, mudflats and sand beaches, can undergo dramatic alterations in extent. The precise nature of these changes are also poorly understood, particularly over the longer term. Ongoing studies of sediment dynamics in the upper reaches of the Bay of Fundy by Amos and his colleagues (see Daborn 1997 and Greenberg et al. 1997 for overviews) are providing many useful insights about the factors contributing to sediment stability, erosion and transport and thus are expanding our predictive capabilities. Reductions in water velocities upstream of the dams can also enhance deposition of fine sediments that smother and effectively destroy the clean gravel spawning beds required by many species of anadromous fish for successful egg development (Dow and Braasch 1996). The magnitude of the ecological impacts of such changes in water flow and sediment dynamics are strongly influenced by local conditions and operational practices. Most dams have been in place for many decades or more, and thus information about prior habitat conditions and the nature of subsequent ecological impacts is largely informed speculation or based on anecdotal reports..

Even when dams do not result in the smothering of fish spawning habitat, they can effectively impede or block the passage of anadromous fish migrating from their marine feeding areas to those critical riverine spawning areas. In such situations there is a "functional loss" of habitat rather than a real physical loss. Although a variety of other stresses have undoubtedly contributed to the serious decline in Atlantic salmon numbers in rivers throughout the Gulf of Maine (Cutting et al. 1994; USFWS 1995), it is likely that dams have also played an important role by causing the actual and functional loss of key spawning habitats. In most New England and Maritime rivers runs of anadromous alosids (principally American shad, Alewife and blueback herring) have also been adversely affected by dams and inadequate fishways (Rulifson 1994), and in Maine, shad runs have been rare since the early 19th century, when impassable dams were built on most large rivers. Dams and other such barriers are also thought to have been the principal cause of the dramatic declines in most other anadromous fish populations in Maine (Fefer et al. 1980), and this is probably true throughout the Gulf region. Some dams incorporate fishways in their design to facilitate the passage of migrating fish. However, studies have shown that even functional fishways prevent the passage of up to 10% of active fish such salmon (Fefer et al. 1980), while less energetic fish such as sturgeon, shad and striped bass probably fare much worse. Many dams throughout the region have no provision for fish passage, while many of those that do are largely ineffective, because of poor design, construction or maintenance (Shelley in press). Clearly there appears to be a critical need for the construction, upgrading or routine maintenance of fishways throughout the Gulf region.

Another option might be to breach or remove some of the dams that have long outlived their usefulness and have largely been abandoned. There are periodic proposals for the removal of such structures, such as on the Presumpscot River flowing into Casco Bay (Peter Milholand pers. comm.). However, there are concerns that such remedial action could initiate a new round of undesirable and largely unpredictable ecological impacts both upstream and downstream. Some of these concerns involve the release of large quantities of trapped sediments, as well as the possible remobilization of long buried contaminants. The short and long-term implications for critical riverine, estuarine and coastal habitats of any such mitigative efforts need to be carefully studied and weighed on a case by case basis before any dams are removed.

On each rising tide, seawater surges into and over many coastal habitats; up creeks and rivers, into and over salt marshes, into protected embayments and through other narrow channels. However, during the past century this natural diurnal seawater movement has been increasingly impeded or diverted in many coastal areas around the Gulf by various , structures across watercourses, channels and wetlands. In every region causeways, road bridges, railroad beds, and other barriers block this tidal exchange, or reduce it by restricting flow to culverts or other narrow channels. In larger centers, such as Boston, Portland and St. John, the land-sea interface has been highly "engineered" over the past century or more and tidal restrictions abound. Even in many seemingly undeveloped estuaries and embayments road crossings, bridge abutments and other structures interfere with tidal flow and destroy or degrade vulnerable habitats. Their design, siting and construction have been determined almost exclusively by community needs, engineering or economic considerations, with little, if any, regard for the potential impacts on the integrity of, and ecological interactions between, the marine habitats the structures bisect.

For the Gulf as a whole there appears to be no comprehensive inventory of the many such structures that interfere with tidal flows, although some regional surveys have been undertaken. For example, half of the 125 tidal crossings in Essex Bay, MA were visually identified as possible sites of significant impedance of tidal flow, and 25 of the sites had a 5 inch or greater difference between the upstream and downstream tidal range. (Mountain et al. 1997). Most of the structures are many decades old and are reportedly poorly designed, sited and maintained (Shelley in press), and few have been evaluated regarding their adequacy in allowing tidal exchange. In almost all cases there is little or no information available about the nature of the impacts on nearby marine habitats or about the areal extent of the habitats affected. Benthic, water column and intertidal mudflat habitats in estuaries and coastal areas may all be directly or indirectly influenced by reductions in tidal exchange. The particular vulnerability of salt marshes to such tidal restrictions is discussed in section 5.2. It has been suggested that the adverse effects on marine habitats of many of these structures could be reduced or eliminated by enhancing tidal exchange by installing new culverts or by enlarging existing ones. (Burdick et al. 1994). However, each situation is unique and remediation needs have to be carefully assessed on a site by site basis.

Harvesting of natural resources has been a mainstay of local economies for centuries throughout the Gulf of Maine. Sometimes the marine habitats adversely affected may be far removed from the actual area of harvesting. Harvesting of resources far inland, as well as along the coast in the intertidal zone or in shallow coastal waters can have significant impacts not only on the resource species being harvested , but on critical wetland and marine habitats as well. In this section we briefly consider the nature and potential marine impacts of resource related activities such as forestry, agriculture, aquaculture, intertidal harvesting, subtidal harvesting as well as mining and submarine aggregate extraction.

forestry

Since the earliest days of European settlement forestry has been a major activity throughout the Gulf of Maine watershed. The history of this lumbering and the widespread impacts that it has had on the landscape of the region has been recently reviewed by Conkling et al. (1995) and Platt et al.(1995a). By the mid nineteenth century between two thirds and three quarters of the forests in the coastal regions had been cleared for settlement and agriculture, and much of the remainder was being intensively harvested to supply fuel and lumber. By this time agriculture in the region had begun to decline in importance and much of the cleared land reverted to forest, particularly around the northern areas of the Gulf. Today, large tracts of New Brunswick, Maine and New Hampshire are heavily harvested "industrial forests" (Conkling 1995) that bear little resemblance to the former mature forest ecosystems that dominated the area. These massive transformations in the natural landscape have over time been accompanied by changes in patterns of freshwater runoff into rivers and estuaries, as well as by increases in soil erosion and in the sediment loads transported into coastal waters. From today’s perspective it is difficult to gauge the extent and the effects of these long ago changes in land cover on estuarine and coastal habitats. Today, forestry activities, along with agricultural practices, still contribute to the elevated sediment loading and hydrologic alterations reported in many rivers and estuaries.

The early forest industry also had other significant impacts on coastal habitats.. The first sawmill was built in 1623 in Berwick, Maine, and over the next two centuries water-powered sawmills sprung up on the banks of virtually every significant river flowing into the Gulf. Many of these involved the earliest efforts to alter river flow patterns with dams (see section 4.3). In addition, the mills released vast amounts of sawdust and bark chips into the river to join the large volumes of wood debris produced by the massive log drives on many of the larger rivers. The pulp mills that eventually began to supplant the saw mills also released large amounts of wood wastes as well as variety of other noxious materials into rivers and estuaries (Eaton et al. 1994). The potential impacts of these wood wastes on benthic habitats are discussed in section 5.8. Although the organic wastes and toxic effluents from modern mills are more carefully regulated, in the Maritimes the many pulp mills still produce more contaminated effluent than any other industrial source in the region (Eaton et al. 1994).

Like forestry, agriculture has a long history in the Gulf of Maine watershed, particularly in the coastal regions which were settled first. The earliest European settlers began diking and draining the vast salt marshes and clearing the great forests of the region in order to expand agricultural production. Only about 15% of the original salt marshes remain in the Bay of Fundy region and less than half remains along much of the remainder of the Gulf of Maine coast (Burdick et al. 1994).. Much of the remainder of the salt marsh has been extensively ditched to facilitate salt hay harvesting or to control mosquitoes. Agricultural activities in the region reached a peak in the mid nineteenth century and subsequently the proportion of land devoted to agriculture declined with the opening up of western farmlands by the railroad and the end of the market for forage by the widespread use of the gasoline engine (Platt et al. 1995a). It is difficult to assess the nature and extent of the impacts of these early agricultural undertakings on the coastal ecosystems, but they were undoubtedly profound. Current trends in agricultural land-use in the region are being monitored as part of the NOAA Coastal Change Analysis Program (C-CAP). Although agriculture now encompasses a much smaller proportion of the landscape (Stevenson and Braasch 1994) it still significantly influences aquatic production, water quality and habitat integrity in riverine, estuarine and coastal waters. The most pronounced effects include contamination of aquatic habitats with an array of pesticides, inorganic nutrients and livestock wastes, as well as increased water runoff from tilled ground with an accompanying increase in soil erosion and sediment loads in rivers and estuaries. In some areas river water is diverted for irrigation and other agricultural purposes. For example, in important blueberry growing areas of Northern Maine substantial amounts of water are drawn from the Pleasant and Narraguagus rivers for growing and processing the crop, altering the flow of these rivers (Shelley in press).

One of the more dramatic phenomena that has taken place in the northern Gulf of Maine over the past two decades has been the rapid growth of the salmonid aquaculture industry. The development of fish farming in the region has been described in some detail (Cook and Black 1993 and Milewski et al. 1997). The industry was launched in southeastern New Brunswick in 1978, and by 1996 there were a total of 77 approved farm sites producing over 16,000 mt of salmon worth almost $120 million (Canadian). Following upon the early successes, the industry subsequently expanded into Nova Scotian waters, particularly the Annapolis Basin, and also into Cobscook Bay and other areas of northern Maine. The southernmost finfish farms at present are in Blue Hill Bay, just south of Mt. Desert Island. Maine (Churchill 1997). The great majority of the sites are located in the protected waters of Passamaquoddy Bay, L'etang Inlet and Cobscook Bay, and in the lee of nearby coastal islands. With most of the suitable inshore sites occupied, and facing conflicts with traditional fisheries and rising environmental concerns (Milewski et al. 1997) there is some interest in the possibility of expanding the industry to larger offshore sites (Harvey in press). Research is also ongoing into the possibility of farming other marine species such as halibut and haddock (Waiwood et al. 1994).

Unlike finfish aquaculture, which is largely confined to the well-mixed areas of the northern Gulf where winter water temperatures are slightly higher, various species of shellfish are being farmed at sites along much of the coast of the Gulf. Only the silt-laden waters at the head of the Bay of Fundy appear to be unsuitable. Species in culture include American and European oysters, mussels, sea scallops and clams in northerly areas. Clam and sea scallop aquaculture is a rapidly growing industry in Cape Cod Bay (Shelley in press). In some areas sea urchins are also held in cages for extended periods until ready for market. Shellfish aquaculture involves restraining or confining the animals but allowing them to feed passively upon natural particulate material in the water. To date there have been few reports of adverse environmental effects from this type of operation.

In contrast, Salmon farming involves confining large numbers of fish in a small area and feeding them intensively with processed food. As a result a large quantity of particulate and dissolved organic waste, comprising uneaten food and feces is continually released in a small area over an extended period. The nature and quantities of the wastes generated by fish farms of various sizes and types have been well documented (see references in Wildish et al. 1993). There is growing concern about the potentially adverse impacts of these wastes on coastal marine habitats and communities (Milewski et al. 1997), and some of these are addressed more fully in section 5.8.

In many areas of the Gulf a variety of marine species are harvested from the intertidal zone. These include, clams and baitworms on mudflats, and mussels, periwinkles, dulse, irish moss and rockweed on rocky shores. There are growing concerns about the potential impacts of local overexploitation on both the resource species themselves as well as on the intertidal communities and habitats that sustain them (Rangeley 1994), particularly as harvests of many of the so-called underutilized species have been expanding rapidly in recent decades as traditional fisheries resources have declined.

Clams, mussels and marine worms have long been commercially important in various parts of the Gulf. It is estimated that in Maine alone the annual value of the fishery for these three species is $13-$15 million (US) (Harvey et al. 1995). Clam, Mya arenaria, digging on intertidal mudflats is unquestionably the most economically important of these activities. The mudflats also yield an abundance of polychaete worms, particularly the bloodworm, Glycera dibranchiata, that are commercially harvested for use as bait in recreational fishing (Klawe and Dickie 1957; Creaser et al. 1983). In recent decades, baitworm populations along much of the eastern seaboard of the U.S. have been greatly reduced by overexploitation and, as a result, much of the harvesting pressure has shifted to the extensive mudflats of the upper Bay of Fundy.

Both clam and baitworm harvesting involve digging up and overturning large volumes of sediment. There is a growing concern about the potential effects of this continuing disturbance of large areas of substrate on the mudflat community as well as upon the stability and integrity of the mudflats themselves. Mudflats are dynamic structures, and studies by Amos and his colleagues (see Daborn 1997 and Greenberg et al. 1997, for overviews) have shown how sensitive the balance between erosional and depositional processes is to a variety of biological and physical factors. At present we know little about the impacts of such repeated human disturbance of large areas of sediment habitat.

On rocky shores, hand harvesting of species of seaweeds such as irish moss, dulse and rockweed has also been carried out for generations, to provide food, fertilizer, soil conditioner or mulch. However, large-scale commercial harvesting of rockweed for fertilizer and alginate extraction has occurred over the past three decades, primarily in Digby and Yarmouth counties of Nova Scotia (Gordon 1994a). Initial seasonal, part-time hand cutting and raking by fishermen were soon replaced by the use of large mechanical harvesters equipped with a vacuum device and cutter blades that could collect far larger quantities of rockweed. The new harvesting technology dramatically increased the harvest in SW Nova Scotia from about 5000 T in 1985 to 27,000 T in 1988 (Sharp and Tremblay 1989) and resulted in the removal 40-80% of the rockweed biomass in harvested areas (Rangeley 1994). There were soon indications that the beds were being overexploited. Since 1990 the harvest has largely reverted to hand cutting and raking (Sharp et al. 1994), and mechanical harvesting has been prohibited in eastern Canada for several years (Glyn Sharp pers. comm.). Increasing pressures by processing companies to further expand the harvest has resulted in the launch of a three year pilot project in the waters of southeastern New Brunswick to collect 10,000 tons of rockweed. Whether the industry will be permitted to expand further in New Brunswick is at present uncertain. In Maine rockweed harvesting has occurred on a much smaller scale, largely for agricultural use and for packing live lobsters (Wippelhauser 1996). However, if the industry is successfully developed in New Brunswick, and if market conditions improve further, then it is likely that there will be considerable pressure to expand the harvest in Maine. There are concerns among scientists and environmental groups that by harvesting rockweed we are, in effect, harvesting important marine habitat. It is increasingly being recognized that rockweed beds are critically important as foraging and refuge areas for a wide variety of marine species, including several commercial fish species. The important ecological roles of rockweed in the coastal ecosystem and the potential consequences of harvesting it are described more fully in section 5.6.

Periwinkles too have long been sustainably harvested by hand for local consumption in many parts of the Gulf. In the Maritimes in recent years the pace of harvesting has increased dramatically with the introduction of suction harvesting techniques. These dredges not only vacuum up most of the periwinkle standing stock in an area, including all size classes, but also take many other intertidal species as well. There are concerns that there may be overexploitation of the resource (Kearney 1994) as well as considerable habitat damage.

The use of heavy trawl gear, and in particular its threat to the integrity and productivity of benthic habitats, has been identified as a problem throughout the Gulf of Maine (Dow and Braasch 1996). National Marine Fisheries Service data reveals that the intensity of such trawling activity has risen sharply in recent decades (NEFSC 1995). It has even been suggested that virtually all seafloor areas of the Gulf are being subjected to some degree of perturbation by heavy fishing gear almost every year (Auster et al. 1995). Some of the potential ecological implications of these activities are discussed in Langton (1994). Although the use of mobile fishing gear is not included in the GPA list of land-based activities, arbitrary exclusion from this report seems inappropriate given growing public concerns about the potential damage being done, particularly to nearshore benthic habitats. It is difficult to argue that such inshore fisheries are not land-based. In coastal areas and estuaries along the entire coast heavy gear has long been used, and is still being used, to harvest groundfish, scallops, urchins and mussels (Shelley in press). For example, in the St. Croix Estuary located on the U.S. - Canada border, each year up to 10 Canadian and 60 American fishing boats drag the area for scallops (Harvey in press). Similar intense fishing pressures are reported in almost every estuary and embayment around the Gulf (Shelley in press). In parts of northern Maine dredging for sea cucumbers is a growing industry. These activities all involve dragging heavy trawls or dredges repeatedly over extensive areas of the sea floor in relatively shallow waters. In some areas the disturbance continues year round as different species are harvested in turn. The repeated passage of such heavy gear tends to level out bottom irregularities and uproot macroalgae and eelgrass. This tends to reduce the structural complexity of the benthic habitat, which could in turn reduce overall biodiversity as specific microhabitats and protective refuges are eliminated. Trawling and dredging also resuspends sediments into the water column, temporarily increasing turbidity and enhancing deposition on nearby habitats. Although the potential for substantial habitat degradation appears great (McAllister and Spiller 1994), the research carried out thus far has not provided definitive answers (Langton 1994; Jones 1992). There is a critical need for more detailed information about large and small scale effects of trawling and dredging (Dow and Braasch 1996).

Quarrying in coastal regions, and aggregate extraction in intertidal zones or offshore areas, are raising concerns about potential effects of the activities on marine habitats in the vicinity. The likely environmental effects of coastal quarries, such as the proposed granite quarry on the shoreline of the St. Croix Estuary (Harvey in press), are at present poorly understood. All around the Gulf, large amounts of sand, gravel and cobble (collectively termed aggregate) are used each year for road construction, building, and other industrial purposes. Up to now, most of this has come from terrestrial sources. Over the past century substantial volumes of aggregate were also extracted from accessible intertidal deposits in various places. The impacts of these activities on beach and subtidal habitats were generally not monitored, but must have been substantial. In all jurisdictions such removal of beach material has been prohibited or carefully regulated for decades (e.g. the Beach Preservation and Protection Act, enacted in Nova Scotia in 1975).

However, convenient terrestrial sources of aggregate are rapidly being depleted and new quarries face increasing public opposition and tougher environmental standards. As a result there is growing interest in the possibility of mining aggregate from submarine sources. In several places around the Gulf promising submarine deposits have already been identified and mapped, such as in Massachusetts Bay (Duane et al. 1988) and Scotts Bay in the Bay of Fundy (Greenberg et al. 1997). Many other deposits around the Gulf are too small or too deep to be mined economically at present. Although aggregate extraction from submarine deposits has not yet been undertaken on a commercial scale in the Gulf, successful operations in Europe and Japan have demonstrated the technical and economic feasibility. The aggregate is usually recovered with a trailer suction dredge. A vessel, equipped with a high capacity pump and a large diameter pipe ending in a large dredge head, moves slowly forward while sucking up bottom deposits. The material is washed through screens to remove mud and the sand and gravel dumped into the hold. When the ship is dredging , a large plume of silt is released. As the ship criss-crosses an area it excavates trenches in the seafloor, up to a half a meter deep and two meters wide. With continued dredging the entire seafloor can be lowered significantly over a large area.

Studies carried out elsewhere suggest that there are a variety of potentially significant environmental impacts that may accompany suction mining of submarine deposits (Day 1994). Seabed mining and navigational dredging (see section 4.2) use similar methods and have similar potential environmental impacts, differing primarily in intent and location. Pearce (1994) reviewed the available literature on both these activities pertaining specifically to the Gulf of Maine region. The principal environmental concerns include direct destruction of bottom habitat and communities, enhancement of coastal erosion and the generation of dense sediment plumes that could conceivably adversely affect primary production, zooplankton, ichthyoplankton, fish and marine mammal populations.

Electrical generating stations are common on many of the larger rivers flowing into the Gulf of Maine as well as at many locations around the coast (Dow and Braasch 1996). Habitat impacts of hydroelectric generating plants situated on rivers are considered in section 4.3 that deals with dams, as it is largely the impoundment of water that leads to habitat modification, although turbine operations do kill fish. The fossil-fuel and nuclear power plants are primarily situated in coastal areas in order to be close to the centers of power demand as well as to sources of water for cooling purposes. While there are various contaminant issues pertaining to both types of power plants (Eaton et al. 1994), it is the release of heated water that may physically alter habitats, and is thus of particular interest here. The heated water released from these plants may form plumes that are as much as 10-15 degrees warmer than the ambient water temperature (Vadas et al. 1976; Kennish 1992). Some of the potentially adverse effects of such thermal plumes on the marine ecosystem are well described in Fefer et al. (1980). The warmed water has been shown to adversely affect rockweed beds near the outfalls (Vadas et al. 1976) and it also alters the thermal characteristics of local benthic and water column habitats. This could influence the use of these habitats as migration, spawning or nursery areas by fish and invertebrates (Eaton et al. 1994). In most cases dispersion of the heated water is relatively rapid and thermal affects are largely confined to within a few hundred meters from the point of discharge. The few studies that have been conducted in the region suggest no major, widespread impacts from the discharges (Eaton et al. 1994).

The causeway across the Annapolis River Estuary in Nova Scotia was originally constructed to protect upstream dikelands from the tidal flooding. In 1984 the structure also assumed a new role as the site for the first operational tidal power generating station in North America. Engineers had long dreamed of harnessing Fundy's surging tides, but it wasn't until the late 1970's, in the wake of rapidly rising crude oil prices, that a major program of economic, engineering and ecological studies was undertaken to assess the feasibility of a proposal for a massive tidal power project somewhere in the Bay. The potential ecological impacts of the proposal were examined in considerable detail in a suite of studies summarized in a variety of publications (Daborn 1987; Gordon and Dadswell 1984; Gordon 1994b), while some of the potential repercussions for the broader Gulf of Maine region are discussed in Campbell (1986) and Greenberg (1984). It is possible that there could be "major unforeseen consequences" for the whole Gulf ecosystem (Kelley and Kelley 1995). Although the full-scale tidal power project was eventually shelved, the 20 megawatt Annapolis facility was eventually constructed as a pilot-scale prototype to evaluate engineering design and environmental impacts.

Passage through the turbines kills a significant proportion of many species of anadromous fish whose populations are already stressed by other factors (Stokesbury and Dadswell 1991). However, the impacts on riverine, estuarine and marine habitats have been more subtle and difficult to discern. Indications are that sediment dynamics, and benthic and mudflat habitats, have been altered both in the river and in the estuary, although it has not been possible to discriminate between the effects of the tidal power plant installation from those of the earlier causeway construction. It has been suggested that an even larger scale Fundy tidal power installation could affect coastal currents, patterns of primary productivity, flushing rates in estuaries, distribution and settling patterns of larval fish and shellfish and the location and magnitude of oceanographic fronts over extensive areas of the Gulf (Kelley and Kelley 1995).

The shorelines and coastal waters of the Gulf offer a wealth of recreational opportunities to the 3.6 million residents of the region. Many millions more visit the coastal area each year to enjoy many of these same activities and amenities (Stevenson and Braasch 1994). Much of the shoreline development in many areas has been a direct result of the steadily increasing growth in tourism. In certain areas, seasonal increases are dramatic; for instance, in Maine, its 4.25 million annual visitors quadruple the states population. Since 95% of these visit the coast, this results in a transient visitation that is seven times the resident population of the coastal counties (Colgan 1989). The same author points out that statistics on tourist populations are questionable as much of the information is either not collected, intentionally withheld or combined with other factors. One of his recommendations to the Gulf of Maine Council was for the systematic compilation of these statistics.

Tourism is the fastest growing business everywhere, and it is predicted that within a few years tourism will be the number one industry world wide. Tourism is more than a business consideration, as it fundamentally alters the natural environment and the social order. It has been shown in other regions that the tourist industry can be a valuable proponent for healthy coastal environments (McNeely and Thorsell 1988; OECD 1980). However, the use of natural resources for outdoor recreation, and tourism can produce conflicts with conservation and protection efforts (Kenchington 1990). Coastal environments can be stressed by tourism and its extensive infrastructure and as well, tourism can be constrained by conservation concerns. Concepts such as carrying capacity are of limited value unless the scientific and cultural issues are addressed (Stankey and McCool 1990).

Coastal islands are particularly attractive to tourists as well as to inhabitants of nearby urban centers interested in a second seasonal residence. On the islands of the Outer Banks of North Carolina, from 1980 to 1990 there was a 100% increase in building permits. The most significant problem was the need for groundwater to sustain growth, which was in direct conflict with the maintenance of wetland quality. Septic contamination of the groundwater and the mining of barrier dunes for fill were other significant conflicts (Miller and Auyong 1990). At Swan's Island ME, the addition of 15 - 20 wells a year caused salt water seepage and a concern for wetland "recharge areas" and on Monhegan Island the aquifer that supplies the 450 residents runs out in a dry summer. Coastal and island communities, whose economies are largely sustained by tourism and seasonal residency, are strongly opposed to initiatives that seek to limit or control tourist numbers.

Sandy beach habitats and the associated dune systems are also particularly vulnerable to developmental pressures as well as to the associated recreational activities. The high economic value of beach-front property and beaches has largely fueled the continuing beach protection and replenishment programs designed to combat erosion. Intense recreational use of many beaches, such as Wells Beach in Maine, may be jeopardizing their use as critical nesting habitat for some seabirds and shorebirds (Shelley in press).

The rapid growth in recreational boating in many areas has had significant consequences for coastal habitats in many areas. The demand for marinas and mooring sites, particularly from central Maine southward has risen dramatically in recent decades. For example, in the Great Bay Estuary of New Hampshire the number of mooring permits increased 5-fold between about 1975 and 1990 (Shelley in press), and the demand continues to rise. Unfortunately, the most desirable marina sites are usually in protected areas of estuaries and sheltered inlets, which are also prime eelgrass and salt marsh habitats. Construction and dredging for marinas and moorings in areas such as Casco Bay has damaged eelgrass beds, while anchoring can also uproot plants and disturb the sediments (Shelley in press). In addition, the propellers of boats maneuvering in shallow waters cut and uproot the plants and resuspend sediments. The turbulent wash from the rapidly proliferating shallow draft power boats, particularly jet skis, are causing erosion of river banks and salt marshes in many estuaries such as the Annisquam River in Massachusetts (Shelley in press). Meanwhile, on the shore, off road vehicles and recreational ATVs are frequently driven over barrier beaches and dunes, destroying vegetation and triggering erosion. This is affecting not only the larger beach/dune systems of Cape Cod but is also a problem on smaller ones such as Mavillette Beach in Nova Scotia.

The processes involved in the destruction or degradation of coastal habitats as a result of human activities discussed above are often subtle and complex. All around are habitats clearly being degraded, but in only rare instances is it possible to confidently point to a "smoking gun" responsible for all the damage. In a very few instances, such as the burying of a saltmarsh with fill, or the dredging of an eelgrass bed, the linkage between cause and effect are readily identifiable and unequivocal. More often, there are a series of complex intermediary processes leading up to the degradation of habitat, and these are typically insidious, gradual and poorly understood. In most situations a variety of confounding factors, such as natural ecosystem cycles or cumulative effects, are also at play and often obscure cause and effect relationships.

Perhaps the most difficult factors to account for are the natural variations that occur in any ecosystem, because an adequate understanding of their effects inevitably requires intensive, long-term studies of the components of the ecosystem of particular interest. Some of these ecological changes are unpredictable, at least on the basis of our current knowledge, and include things like sudden population explosions of particularly influential species or the sudden outbreak of diseases. For example, the effects of rapid increases and abrupt decreases in sea urchin populations on the stability of the kelp bed habitat in the Gulf is described in section 5.11, while the dramatic impacts of wasting disease on eelgrass meadows, and on the biota that depend on them, is discussed in section 5.10.

Many of the ecosystem changes are cyclic in nature, often linked to solar or lunar periodicity. An example is the 18.6 year cycle in tidal amplitude (nodal cycle) associated with variations in the declination of the moon’s orbit (Greenberg 1983). There are indications that the nodal cycle may be associated with variations in ocean currents, vertical mixing and water temperature (Loder and Garrett 1978), which in turn may influence ecological processes and populations of marine organisms inhabiting the pelagic zone (Cabilio et al. 1987). The exact nature of the links in this complex chain of events are only poorly understood.

Other ecosystem changes, such as sea level rise, are progressive in nature, at least from our limited time perspective. The Gulf is relatively young as an oceanographic and ecological entity, released from the grip of great glaciers little more than 13,000 years ago (Kelley et al. 1995). Early on, the bounding banks rose above sea level, making the Gulf an almost enclosed shallow estuarine basin draining into the Atlantic. Rising sea level has submerged the offshore barriers and allowed the sea to encroach up river valleys and expose new areas of shoreline to the forces of erosion. Gradual crustal subsidence in the region has caused sea level to slowly rise over the past 7,000 years and it is now estimated to be rising at a rate of about 13-21 cm each century (Scott and Greenberg 1983), exacerbated to an as yet unknown degree by melting of polar ice caps as a result of global warming. Some of the possible impacts of such sea level changes on production and on coastal habitats in the Bay of Fundy - Gulf of Maine region are described in Gordon (1986). Almost all coastal habitats are probably undergoing some changes associated with rising sea level and increasing tidal amplitude, although the exact nature of these responses are not well understood at present.

Interpretation of the underlying causes of habitat changes is further complicated by the fact that many different human activities are contributing to the environmental problem simultaneously. This may reflect the inputs from many different sources of a single ecological stressor. For example, mudflats and salt marshes are vulnerable to any activities that alter sediment dynamics in the watershed or coastal zone. It is well known that agriculture, forestry, road construction, urban development, coastal armoring, dam construction, dredging and many other human activities can all result in modifications to patterns of sediment erosion, transport and deposition and thus to the alteration of habitats dependent on sedimentary processes. In many areas habitat changes probably reflect the combined results of many different ecological stresses acting simultaneously, the so called cumulative effects. For example, an eelgrass bed may be being systematically degraded by a combination of boating activity, nearby dredging, excessive nutrient loading and coastal infilling. Occasionally the impacts of these combined stresses can be far greater than would be expected from simply adding the effects of the individual stresses, the so called synergistic effect. Most often, a degrading habitat reflects the combined influences of cyclical processes, progressive environmental changes and multiple human activities all working together.

Perhaps the biggest obstacle to attempts to link habitat degradation to specific causes is the general inadequacy of our present understanding of many of the complex physical, chemical and ecological processes involved. Typically, our level of understanding and predictive capabilities decline rapidly with increasing spatial and temporal scales, and usually it is the impacts over wide areas and over the longer-term that are the greatest threat to overall habitat integrity and ecosystem health. For example, there have been many studies done on the impacts of finfish aquaculture pens on the benthic habitat immediately below them, and a number of useful models have been developed to describe the process. Although there are varying degrees of localized habitat degradation (see sections 4.4 and 5.8) there appears to an acceptance in some quarters that the impacted areas are small and the overall ecological consequences are thus acceptable. However, we know very little about the transport of the so called mariculture sludge away from the cage sites, where else and how it might accumulate and what the ecological impacts of this accumulation might be. There are concerns that broad areas of intertidal and subtidal habitat could be degraded over the long-term by aquaculture wastes accumulating faster than they can be assimilated by the system. Similarly, although progress is being made in understanding some of the processes that effect sediment dynamics over small areas in the short-term (see section 5.3), many of the longer-term processes involving large geographic areas still elude us, even though these may be the most worrisome with respect to significant habitat degradation. For example, some scientists suspect that some of the adverse changes now being seen in many mudflat habitats in the upper Bay of Fundy may be the long-delayed, or slowly building, effects of alterations in sediment dynamics that were triggered by the damming of major rivers in the region over three decades ago (Brylinsky et al. 1997).

There are often a number of intermediary processes intervening between the actual land-based activity and the ultimate impact on a marine habitat. Mostly these are complex and poorly understood, leading to further difficulties in linking anthropogenic causes and ecological effects. Often too, several human activities feed into the same intermediary processes and several different habitats may experience the brunt of their impacts, adding yet more variables to an already complex situation. Although there are several of these intermediary processes that could be described, perhaps the most important from the present perspective are eutrophication, sediment dynamics and hydrology.

Sediment dynamics

Another important intermediary factor that often complicates the linkages between land-based activities and habitat impacts is sediment dynamics. The structure and function of many coastal habitats may be markedly influenced by alterations in the processes associated with sediment erosion, transport and deposition. In fact, according to Daborn (1991), in the upper Bay of Fundy in particular, it is the dynamics of the sediments that are the key to understanding the functioning of its ecosystems. This is probably true to a lesser degree for many of the coastal habitats around the Gulf of Maine. As we have seen, a great many land-based activities can influence sediment dynamics. Considerable progress is being made in understanding and modeling some of these sediment related processes (see Daborn 1997 and Greenberg et al. 1997 for overviews). The development of the concepts of a dynamic balance between sediments and water and of an "equilibrium capacity" (Amos 1995) has been particularly useful in understanding the effects of human modifications to estuaries (Daborn 1997). Thus water bearing less than its "equilibrium capacity" of sediment will tend to mobilize sediments from nearby shorelines, while water with more than this will tend to deposit sediments in areas that will accommodate them. Similarly, a multidisciplinary project involving over 30 scientists from 5 nations, the Littoral Investigation of sediment Properties (LISP) is providing a better understanding of the behavior of fine sediments in coastal waters and aiding in the development of improved sediment dynamics models (Daborn 1991). In spite of these and other recent advances in knowledge it is still difficult if not impossible to predict the impacts of many land-based activities on sedimentary processes, particularly over larger areas and in the longer term. Further discussions of specific sediment issues are included in the sections 5.3 and 5.10.

Closely linked to the question of sediment dynamics is another important intermediary factor, namely hydrological processes. Most sediments are mobilized and transported by water, so anything that influences the volumes or patterns of water flow or the velocity or direction of currents will also have a profound effect on sediment dynamics. In addition, altering the patterns of water flow in rivers by the construction of dams can significantly change a range of other factors such as water temperature, salinity, nutrient loading and mixing in the estuaries into which they flow, and in nearby coastal areas. As we have seen many of the land-based activities described above have some influence on water flow characteristics and thus the potential to alter habitats. Often, many different activities in a watershed influence hydrodynamic processes in nearby rivers, estuaries and coastal areas in a complex cumulative or synergistic manner. Many of these hydrologic processes are poorly understood, making it difficult to predict effects on coastal habitats. A few of the more important of these hydrologic issues are dealt with in sections 4.3 , 5.2 and 5.3.

The Gulf of Maine is a well defined and ecologically distinctive (Conkling 1995) embayment encompassing some 90,700 km² (Stevenson and Braasch 1994) of the continental shelf of eastern North America. Extending from 41^o N to 46^o N and longitudinally from 65^o E to 71^o E, it is bordered by a 5,600 km coastline involving the states of Massachusetts, New Hampshire and Maine and the provinces of New Brunswick and Nova Scotia. Its 90,700 km² watershed drains substantial proportions (36 -100%) of each of these territories as well as a small area of southern Quebec (GOMCME 1992). It is a shelf sea with an average depth of 150 m and a maximum depth of 377 m (Kelley et al. 1995). It includes three large basins that exceed 250 m (Townsend 1996) and is bounded to seaward by four large banks less than 60 m deep (Lynch 1996). These reduce the influence of the adjacent Northwest Atlantic Ocean, creating a relatively enclosed oceanographic system, with a well defined internal circulation pattern, that can be considered a distinctive ecological entity. However, the oceanographic isolation is not complete, because three major channels connect the Gulf with the Atlantic (Lynch 1996). The northernmost of these is the main entry for Atlantic water from the Scotian Shelf, while the deeper Northeast Channel between Browns and Georges banks permits the only inflow of nutrient-rich slope water (Kelley et al. 1995). The Great South Channel southeast of Cape Cod is the main outflow from the Gulf onto the adjacent continental shelf (Yentsch et al. 1995).

The coastline of the Gulf is complex and varied. This can largely be attributed to the eight distinctive geological regions along the coast (Kelley et al. 1995), characterized by rocks of differing origin, composition, orientation and resistance to erosion. Thus, there are the readily erodible sandstones and shales that surround the upper Bay of Fundy, the stark resistant bedrock headlands that characterize much of the Maine coast and the southernmost area towards Cape Cod that is completely lacking in bedrock and comprised largely of glacial deposits of sand and gravel (Kelley et al. 1995), and other more subtle geological distinctions elsewhere. The variability in oceanographic conditions and in the coastal and submarine geology and topography results in a great diversity of habitats within the Gulf of Maine.

This report adopts a rather general definition of the term habitat. Typically habitat is defined from the perspective of a species and its preferred portion of the environment. However, for any given species this encompasses many different parts of the environment that can vary with age, life history stage, season or even time of day. Typically it include spawning grounds, nursery areas, feeding zones and migration routes (Gordon 1989). In this context, identification and ranking of important habitats requires a judgment about the relative value of different species. This is the approach being taken by the US Fish and Wildlife Service and the Gulf of Maine Council in the Gulf of Maine project, where some 161 species of marine plants and animals have been ranked according to a suite of up to 10 ecological, economic or conservation criteria (USFWS/GOMCME 1994). The intent is that this ranked array "can be used as a focus for identifying habitats".

The definition of habitat of Ryder and Kerr (1989) that views habitat as a structural component of the environment that serves as a centre of biological activity is more manageable and suited to present purposes. This approach was used by the Maine Department of Conservation in its classification of ecosystems and natural communities (Gawler 1991), although the marine and estuarine benthic areas alone were subdivided into 67 distinct habitats according to a variety of topographic, geological, physical and chemical factors (Brown 1993). This scoping paper considers only 12 very broad habitat categories; namely, saltmarsh, mudflat, sand beach/dune, cobble beach, rockweed bed, coastal islands, inshore (subtidal) benthic, inshore pelagic, eelgrass beds, kelp beds, offshore benthic and offshore pelagic. The term "pelagic" is used in the sense of Vernberg and Vernberg (1970) to refer to the water column itself, with the inshore pelagic extending from the low tide line to an arbitrary depth of about 50 m, and the offshore pelagic comprising everything deeper. The selected habitats differ considerably in primary productivity as shown in Figure 1.

An estuary is not a habitat as such, but rather a of grouping of several interdependent habitats existing in close proximity. These individual habitats, which are not necessarily restricted to estuaries, are dealt with in the sections that follow. Nevertheless, estuaries are important ecological units that are widely distributed around the Gulf of Maine, and many research programs and conservation activities are focused on estuaries as a whole rather than on the component habitats. This brief overview is included to present some of the habitat issues in an estuarine context.

Estuaries are coastal ecosystems where freshwater flowing from inland rivers meets and mixes with seawater. They are considered to be one of the most productive aquatic ecosystems in the world and are particularly noted as important nursery grounds for many commercially and recreationally important species, including both finfish and shellfish. They are important staging areas for waterfowl and shorebirds. The high biological productivity typical of most estuaries is due largely to the unique hydrodynamics that results when freshwater and sea water mix. A complex set of factors interacts to create a system that concentrates both inorganic and organic nutrients as well as planktonic organisms. In addition, nutrients tend to be recycled rapidly as a result of both physical and biological processes.

The biological communities present in most estuaries are typically characterized as having high biomass but low diversity, the latter being the result of the stress imposed by the high spatial and temporal variations in salinity which make it a difficult environment for organism that have little ability to osmoregulate. The pelagic community of Gulf of Maine estuaries is dominated by an assemblage of diatoms, flagellates, copepods and meroplanktonic larvae. The benthic subtidal community consists mainly of various polychaetes, mollusks and crustaceans, many of which are important food items of commercially and recreationally valuable species. Salt marshes, mudflats and seagrass meadows are other important habitats that are common in many estuaries and are described in more detail in the following sections.

Salt marshes are an important feature of many estuaries and relatively sheltered, low energy, coastal locations in temperate latitudes. They commonly develop on stable or emerging shorelines, but can also be found on submerging coasts if the rate of sediment supply is greater than the rate at which the land is sinking or local sea level is rising. They are highly dynamic systems, responding to the interactions between fresh water, sea water and sediments, growing and receding as these forces change over time. This dynamic response occurs through the growth of the dominant species of vegetation, but is critically influenced by sediment supply, wave action, and human activities. Remnants of once extensive salt marshes are present in various parts of the Gulf, particularly on Cape Cod, along the northern Massachusetts coast, bordering Maine’s extensive and numerous estuaries and in the Bay of Fundy, near its mouth and particularly along the margins of Chignecto Bay and Minas Basin (Gordon 1994a). It has been estimated that there are only about 158 square kilometers of salt marsh remaining in the Gulf of Maine region, with two thirds in the United States and one third in Canada (Jacobson et al. 1987).

In the Bay of Fundy, salt marshes developed most extensively in the inner portions of the system, in Minas Basin, Cumberland Basin and Shepody Bay, and in the Annapolis Basin. Smaller marshes are present in many estuarine locations around the Bay. Typically, the lowest point of a marsh corresponds to the mean high water mark of neap tides , below this the intertidal zone is a mud or sand flat, which is covered by water at least half of the time during any tidal cycle. The area of marsh below mean high water that is flooded regularly is characterized as low marsh. The uppermost level of the marsh (high marsh) extends above the high water spring tide level, where it may be flooded on only a few tides each year. Although the difference in elevation between the upper and lowermost parts of the marsh may be small, variations in flooding depth and frequency, in the influence of fresh water, and in the amount of sediment carried by flood waters, leads to a more or less distinct zonation of different plants.

The marshes are dominated by species of Spartina: S. alterniflora (marsh or smooth cordgrass) at the lower levels, and S. patens (marsh hay) at the upper end. These dominants are interspersed with other plants, notably Distichlis sp. and Juncus sp. (Black grass), Sea lavender (Limonium sp.), plantain (Plantago sp.) and glasswort (Salicornia sp.). Cordgrass is the initial colonizer, establishing itself on new deposits of sediment when these accumulate to the neap tide high water level. They develop extensive root systems that stabilize the sediment, and trap more sediment each time the tide floods the marsh. Much of the new sediment is deposited at the edges of tidal creeks, building up levees upon which the S. alterniflora grows relatively tall. Behind these channel edges the marsh surface is lower, a shorter form of cordgrass occurs. Because of the levees, the marsh is often flooded less regularly, and may contain ponds that harbor a wide variety of invertebrates and some fish. At higher elevations, cordgrass gives way to marsh hay, so-called because it was harvested for cattle feed in earlier times. The marsh and its ponds harbor a limited diversity of animals (e.g., mud snails, polychaetes and some insects), but these are sometimes very numerous, providing a substantial food source for fish (e.g., Atlantic silverside, killifish, mummichog, and three-spined stickleback), and for resident and migratory birds (e.g., swallows, herons, godwits).

Winter ice conditions make the Fundy marshes very different from those further south. Ice 'plucking' during winter can remove much of the above-ground S. alterniflora, adding to the detritus chain. The two Spartina species also exhibit different growth patterns in summer : according to research carried out in Minas Basin, S. alterniflora loses leaves on a regular basis as it grows during the summer, whereas the previous years' above ground growth remains beneath the marsh hay (S. patens), decaying only slowly. Very little of the marsh production is consumed by animals while it is alive, but Spartina detritus is a major contributor to food chains after the leaves have died. The salt marshes of New England differ from those of Fundy in a number of other characteristics, particularly in their much higher organic content (Nixon 1982). They have largely developed on layers of marine peat behind barrier beaches or in river mouths, while those of the upper Fundy region lay upon thick deposits of marine silt (Gordon 1986). The Fundy salt marshes are not quite as productive as those further south. Their role in stabilizing intertidal sediments, providing feeding and spawning habitat for some fish and birds, and contributing to the food chains of adjacent waters was probably of greater significance prior to European settlement (Gordon 1986). The remnant salt marshes that rim much of the Gulf still fulfill many important ecological roles and their structural and functional components are schematically presented in Burdick et al. (1994). Their role in contributing organic matter to other coastal habitats is detailed in Gordon et al. (1985) and Gordon and Cranford (1994).

Coastal salt marshes were once a much more important natural feature of the Bay of Fundy, covering an estimated 28,000 ha. Since the settlement of Europeans, however, these marshes have been extensively diked and converted for agricultural use, leaving only about 6,500 ha (23%) still 'out to sea'. In a number of places, modifications to water flow have caused relatively large salt marshes to develop anew (e.g., as a result of building dams at Windsor, N.S. and Moncton, N.B., and along the seaward side of dikes that are still maintained). In other cases, restrictions of water flow onto marshes resulting from the building of culverts beneath roads have converted marshes that were previously net exporters of organic matter into 'sinks' because they trap floating marsh grass and seaweeds brought in by the tide. The overall effect of these changes has probably been to reduce the relative importance of saltmarsh production to coastal ecosystems.

'Reclamation' of salt marshes was carried out without recognition of their potential role in coastal food webs, and yielded high quality farmland that has been an important part of development in the region. In recent years, however, collapse of hay markets and the high cost of installing effective drainage has somewhat tarnished the image of farmed marshlands, despite demonstrations that when appropriately managed they produce the highest yields in the region. Consequently, they have been considered for other uses, including landfills, industrial, commercial and residential development. In many areas around the Bay of Fundy extensive areas of reclaimed salt marsh that are no longer deemed necessary for agricultural are being converted to freshwater impoundments to provide waterfowl habitat. Ducks Unlimited Canada in conjunction with the Eastern Habitat Joint Venture Program have been spearheading these efforts. Some 100 wetland impoundments in the Gulf of Maine watershed, encompassing over 8,100 hectares have already been developed (Gulf of Maine Times, Vol.2 #1), including the 320 hectare Belleisle Marsh on the Annapolis River Estuary. This trend is likely to intensify in the future. Little if any consideration appears to have been given to the feasibility or potential ecological benefits to the nearby marine ecosystem of restoring these areas as functioning salt marshes. Their important role in support of aquatic food chains, migratory fish and birds, and as habitat for a number of rare or endemic species has generally been ignored.

A salt marsh is an integral part of, and is intimately linked to, the adjacent marine ecosystem. On each tide it receives sediments and nutrients from the sea and returns an abundance of organic matter to nearby coastal waters. In many parts of the Gulf these important exchanges have been impaired by the presence of a wide range of tidal restrictions as outlined in section 4.3. For example, in coastal New Hampshire some 50 tidal restrictions have been identified that affect 20% of the remaining salt marsh (Shelley in press). The trophic consequences of this reduced interaction for the marshes and the adjacent marine habitats are poorly understood (Burdick et al. 1994). Such