IMPACTS OF MANAGEMENT ON FISHERIES DIVERSITY AS A FACTOR AFFECTING SOCIAL RESILIENCE IN COMMUNITIES OF THE SOUTHERN NEW ENGLAND REGION

The present research aims at examining potential impacts of fisheries management on the diversity of species exploited by fishermen in Southern New England, and examining the relationship between this diversity and the resilience of fishermen and, consequently, communities in adapting to changes in their social and physical environments. One way fishermen are able to adapt to environmental and socio-economic changes is through diversification of the catch. The current study argues that certain management plans, especially those focused on limited entry strategies, have been reducing fishermen’s adaptive flexibility, therefore reducing fishermen’s resilience and that of their communities. Two hypotheses were developed: 1) the diversity of species landed by fishing vessels homeported in the studied communities has decreased through time as a consequence of management practices; and 2) A decrease in diversity and flexibility in the fisheries has the potential to negatively affect fishermen’s individual well-being as well as their resilience to changes in the fishery. In order to test these hypotheses, two research methods were used: structured surveys and analyses of secondary data of fishery landings for the ports studied. A total of 157 interviews were conducted with fishermen from the ports of Point Judith, Rhode Island, New Bedford and Fairhaven, Massachusetts, and the Cape Cod region of Massachusetts, more specifically the ports of Chatham, Harwich Port, Hyannis, and Provincetown. Results of analyses show that fishermen perceived a significant reduction in fishery diversity to have occurred as a result of regulations, despite the fact that analyses involving landings data showed a significant decline in diversity only for New Bedford and a slight declining trend for New Bedford and Cape Cod when trawl gear landings were analyzed separately. The latter results can be in part explained by the analyses of landings fluctuation for some of the most important species in the studied region, suggesting that diversification has occurred, possibly as a means of adaptation to change. Qualitative data obtained from surveys suggest that impacts on fishery diversity have a negative effect on fishermen’s resilience. Results from this study have the potential to contribute to the enhancement of knowledge and stimulate important future research about aspects of adaptability in fishing communities specifically with regard to their impact on future policy strategies.


IMPACTS OF MANAGEMENT ON FISHERIES DIVERSITY AS
argues that certain management plans, especially those focused on limited entry strategies, have been reducing fishermen's adaptive flexibility, therefore reducing fishermen's resilience and that of their communities. Two hypotheses were developed: 1) the diversity of species landed by fishing vessels homeported in the studied communities has decreased through time as a consequence of management practices; and 2) A decrease in diversity and flexibility in the fisheries has the potential to negatively affect fishermen's individual well-being as well as their resilience to changes in the fishery. In order to test these hypotheses, two research methods were used: structured surveys and analyses of secondary data of fishery landings for the ports studied. A total of 157 interviews were conducted with fishermen from the ports of Point Judith, Rhode Island, New Bedford and Fairhaven, Massachusetts, and the Cape Cod region of Massachusetts, more specifically the ports of Chatham, Harwich Port, Hyannis, and Provincetown. Results of analyses show that fishermen perceived a significant reduction in fishery diversity to have occurred as a result of regulations, despite the fact that analyses involving landings data showed a significant decline in diversity only for New Bedford and a slight declining trend for New Bedford and Cape Cod when trawl gear landings were analyzed separately. The latter results can be in part explained by the analyses of landings fluctuation for some of the most important species in the studied region, suggesting that diversification has occurred, possibly as a means of adaptation to change. Qualitative data obtained from surveys suggest that impacts on fishery diversity have a negative effect on fishermen's resilience. Results from this study have the potential to contribute to the enhancement of knowledge and stimulate important future research about aspects of adaptability in fishing communities specifically with regard to their impact on future policy strategies.
ACKNOWLEDGEMENTS I would like to express my utmost gratitude to my advisor Dr. Richard Pollnac.
During the five years that I have been his student, he has given me immensurable support and useful advice, and has also become a very important person in my life. I owe so much to his belief in me. P, I admire you professionally and personally and I have learned a lot from you. Thank you. A big thank you to Angela, whom without I would not have been able to finish fieldwork on time, for making those mornings at the docks a lot more fun and hilarious at times. Thanks! I would like to thank Changhua for sharing her experiences as a Marine Affairs PhD student with me and for answering so many of my silly questions. Xie Xie.

Theoretical background and statement of the problem
The past few decades have been marked by an increased emphasis on the importance of understanding social impacts of management in fishing communities (see Colburn et al. 2006 for historical overview). In 1996, the enactment of the Sustainable Fisheries Act during reauthorization of the Magnuson-Stevens Act, implemented National Standard 8 (NS8), calling for the consideration of community impacts and emphasizing the importance of fishing resources for communities (Magnuson Fishery Conservation and Management Act of 1996). This initiative contributed to the development of social impact assessments for fisheries management plans as well as to an increase in research effort in the field of fisheries social sciences with regard to the impacts of management on coastal communities (Olson 2006, Sharp and Lach 2003). According to Colburn et al. (2006) the passage of NS8 required the body governing fisheries in the U.S. to "consider how fishing-dependent communities can adapt and sustain their engagement in marine resources harvesting and processing in the face of complex pressures" (2006: 234).
There is enough evidence in the literature to support the idea that knowledge concerning aspects of community and individual resilience gives managers the opportunity to foresee potential consequences of external change events (pressure), such as new policies, and thus choose options that balance social and economic costs with sustainability goals (see Marshall andMarshall 2007, Coulthard 2012). For that reason, there have been increased theoretical efforts to apply the concept of resilience to fisheries management practices in an attempt to maximize the ability of fishing communities to adapt and deal with change (Adger 2000, Carpenter et al. 2001. Fishing communities in the Northeast Region of the U.S. have been the focus of intense and complex changes in the past few decades. Fish stock declines have resulted in the implementation of strict and intricate regulations, often the cause of frustration and conflict between fishermen and policy-makers. During the past decade the increased popularity of limited access programs in the fisheries led to dramatic structural changes in the fishing industry and consequently in traditional fishing communities. One way fishermen can adapt to changes (e.g. stock and market fluctuations, management restrictions) is by diversifying their fishing activity. This idea is supported by a great deal of research, especially in the fields of 'portfolio,' 'livelihood,' and 'risk management' (Allison and Ellis 2001, Marschke and Berkes 2006, Béné 2009, Ha and van Dijk 2013, Cinner et al. 2010, Sethi 2010, Kasperski and Holland 2013. A diverse catch composition is also believed to be related to better conservation practices in the fishery (Murray et al. 2010).
The means and ability to exploit an array of different resources at different times of the year constitutes part of the adaptive tools fishermen have to overcome change and guarantee a stable flow of income. In other words, diversity of catch functions to maintain adaptive flexibility and help ensure long term survival in the face of adversities. This idea is especially relevant in the context of fishing communities because aspects of job satisfaction among fishermen have been shown to create reluctance to leave the occupation even when income is minimal (Pollnac et al. 2001, Pollnac and Poggie 2008, reinforcing the need for 'withinfishery diversification' (e.g. diversify species, gear, and fishing grounds) as opposed to 'out-fishery diversification' (i.e. seek alternative occupation) (Ha and van Djik 2013

Hypothesis II:
A decrease in diversity and flexibility to exploit different species in the fisheries has the potential to negatively affect fishermen's individual well-being as well as their resilience to changes in the fishery. Impacts on resilience are expected to affect more significantly multispecies fishermen.

Point Judith, Rhode Island
The port of Point Judith (41°23'59"N 71°30'23"W), also known as Galilee, is located on Point Judith Cape, in the town of Narragansett, Rhode Island. The port is located in the village of Galilee, which borders the village of Point Judith to the east and the village of Jerusalem, home to a small fishing port, to the west across the Point Judith channel (figure 1). involved the construction of a "harbor of refuge," that consisted in the placement of three extensive breakwaters totaling more than 11,800 feet (Gersuny and Poggie 1974).
The other large intervention that contributed to the success of Point Judith was the dredging of the channel into Point Judith Pond. Shifting sand deposits at the pond's outlet to the sea obstructed navigation, limiting the access of large steamship boats into the harbor. Although the motivation for improving the port was not primarily related to the fishing industry, impacts of the two interventions on the productivity of the local fisheries were evident: records show that landings increased from 300 tons in 1895 to 3000 tons in 1935 (Gersuny and Poggie 1973).
In 1935, major improvements were made to the Port of Galilee. The government expended $300,000 U.S. dollars with the construction of two piers and the dredging of a thirty-five acre anchorage basin inside the pond, in a project which partially aimed at providing jobs during difficult times for the U.S. economy.
Subsequent to these improvements, fish landings increased drastically reaching a total of 17,000 tons in 1945 (Gersuny and Poggie 1974). The fishing industry continued to prosper in the following decades, and despite capital shortage associated with the World War II, both tonnage and value of catch continued to rise, reaching a record of 142 million pounds in 1957 (Gersuny and Poggie 1974).
The extremely large landings observed for the periods that followed the early 1930s can be associated with the introduction of the first large trawlers in the New England fisheries. As inshore menhaden stocks plummeted, technologies allowing efficient pursuit of fish farther from shore became available. The otter trawler became the primary method in the early 1930s, employed in the capture of mostly whiting and red hake (Hall-Arber et al. 2001).
In 1947 the Point Judith Fishermen's Cooperative Association was created as a means to look out for the economic interests of the local fishermen. As the fleet grew and transformed from a shore fishery into a vessel fishery, private interests in the receiving and transporting of the catch subjected the fishermen to manipulation by the fish-market middlemen who charged the highest possible prices to handle the catch.
Therefore, in order to protect their interests, sixty-five local fishermen formed the cooperative that eliminated the need for the outside middlemen and put the fishermen at an advantage in the market for fish products (Marshall 1973).
Although the cooperative ensured higher fish prices and therefore higher incomes for the fishermen, gross landings of many important species began to decline after 1960. As a response to this decline, Rhode Island fisheries went through a phase of diversification in species, in an effort to maintain a relatively stable amount and value of total landings. As part of this diversification process, offshore lobstering was introduced in the early 1960s (Doeringer et al. 1986 databases show a total of 138 vessels homeported in Narragansett. After the end of the whaling era the city of New Bedford invested in textile manufacturing, which remained its major economic activity and attracted thousands of people to the city until the end of the textile period in the 1940s. After the decline of the textile industry, New Bedford faced a period of severe unemployment but has greatly diversified its economy since then. One way New Bedford responded to this period of crisis was through reinforcing its connection to the sea. Despite the fact that a commercial fishing fleet had been operating in New Bedford since the mid 1800s, it was in the early 20 th century that motors, trucks, and modern refrigeration transformed the industry, and New Bedford soon became a major fresh fish processing center and scallop port on the Northeast Atlantic (Moss and Terkla 1985).
These changes transformed the fishing industry in New Bedford, and fishing vessels no longer had to sail to New York to sell their catch. This allowed for a local competitive market, and the implementation of the fish auction in 1941, which regularized the selling of fish using set rules and time limits. However, as a result of a disagreement between fishermen and boat owners, the auction closed in 1985. Private sales took over until 1994, when the Whaling City Seafood Display Auction was established (Orleans et al. 2010). A display auction allows the buyers to see the catch that is off-loaded into large coolers. Since 1997 the auctions have been performed completely electronically (Whaling City Seafood Display Auction nd).
Once New Bedford was well established as a fishing port, many vessel owners and captains began to buy property in the neighboring town of Fairhaven. Today, while most of the fishing activity is located in New Bedford, Fairhaven has a substantial number of marine service establishments. In fact, the ports of New Bedford and Fairhaven are considered to function as a unit (Portman et al. 2011  Although cod is still an important resource, the fishing activity in the Cape Cod region is characterized by a diverse fleet targeting a variety of different species of fish and shellfish using many different gear types and methods of fishing. Table 3 shows volume in pounds and value for significant species landed in 2012 at the ports  American plaice, Atlantic halibut, redfish, ocean pout, and white hake ** Silver hake (whiting), red hake, and offshore hake

Significance of the study
The present research examines the relationship between fisheries diversity, management, and resilience of fishing communities in the Northeast Region of the United States. One way fishermen are able to adapt to environmental and socio-economic changes is through diversification of the catch. A diverse catch composition is also believed to be related to better conservation practices in the fishery. The current study argues that certain management plans, especially those focused on limited entry strategies, have been reducing fishermen's adaptive flexibility, therefore reducing fishermen's resilience and that of their communities. It is also argued that, a less diversified fishery can have a negative impact on both the ecological resilience of the fish populations and their habitat. Investigating interactions between these phenomena has the potential to make important contributions to a better understanding of relationships between fishermen, their resilience, and the health of the ecosystems they depend upon.
Through an examination of the relationships between changes in catch composition, resilience, and fishery management, this study will contribute to the understanding of adaptability of U.S. fishermen to a changing social and ecological environment. The New England region has been the focus of frequent changes in fishery regulations in response to fluctuations in fish stocks, which strengthens the significance of this research in the studied area. This study will also make a contribution to the pool of knowledge of social resilience in general because it is an attempt to operationalize this variable and show the interactions between resilience, adaptive flexibility, and management -subjects currently of importance to the fisheries scientific community.
Moreover, results from this study have the potential to contribute to the enhancement of knowledge about aspects of adaptability in fishing communities specifically with regard to their impact on future policy strategies. Understanding the links between fisheries diversity, resilience, and management has the potential to contribute to the development of regulations that maximize sustainability and resilience in coastal communities through a focus on adaptive flexibility. The theoretical focus of the present study is believed crucial to the incorporation of the findings of this research into future policy plans.

LITERATURE REVIEW
Chapter II provides a literature review on the main elements and frameworks related to the present study. The first sections of this chapter cover an overview of resilience theory and its application from ecological sciences onto social sciences and, more specifically, onto studies of fishing communities. The following sections cover the literature on adaptive flexibility and diversification in the context of fisheries. This chapter also provides a historical overview of fisheries management strategies implemented in the New England region for two of the main fishery stocks (groundfish and scallops) that are of particular interest to the ports studied in the present research. This section has the objective of providing examples of the potential complexity and intricacy that fishery management processes involve as well as background information that will help to understand the questions being investigated in this study.
Chapter II has the overall objective of providing the theoretical background that led to the hypotheses presented in Chapter I as well as defining the conceptual framework used as the basis for the present study.

Resilience theory 1
The past decade was marked by increasing awareness of the importance of management strategies that consider and maximize resilience in coastal communities (Adger 2000, Carpenter et al. 2001, Cinner et al. 2012).
The resilience perspective emerged from the ecological sciences and marked a switch from traditional stability theories to the idea of ecosystems as having multiple-stable states, or in other words, presenting multiple possible states of equilibrium (Lewontin 1969, Holling 1973. The first ecological studies on resilience focused mainly on interactions between predators and prey and their responses to habitat stability (Holling 1961, Morris 1963, Rosenzweig 1971, May 1972. Ecologists adopted the term resilience from mathematical sciences, where it was originally used to describe the behavior of systems with a single domain of attraction, which can be described as the mode a system tends to be in the absence of disturbance. The level of resilience of linear systems is defined by resistance and speed of return to the domain of attraction of a given system after external disturbances. If the system breaks the resilience threshold, the likelihood that it will return to that previous state is decreased (O'Neill et al. 1986, Pimm 1984. This definition of resilience is often called 'engineering resilience' and it is similar to what ecologists call 'stability' but differs conceptually from the definition of 'ecological resilience' (Holling 1996).
In one of the most renowned publications on ecological resilience, Holling (1973) defines the concept as the persistence of ecosystems and their ability to absorb change and disturbance and still persist or maintain the same relationships between populations and state variables. This view of resilience, based on the idea of ecosystems as having multiple domains of attraction, emerged as a consequence of a concern for a more realistic understanding of natural systems highly governed by uncertainty, as opposed to artificial and deterministic views and models. According to Holling (1973) the important focus when studying complex systems is not how stable they are but how likely it is for the system to move from one domain to another and still persist. A system can have multiple desirable states, and shifts between such states may be indispensable for the existence of the system as a whole. In this sense, systems can be highly unstable and present enormous resilience (Holling 1973). Under the multi-stable state perspective, adaptation is key to maintaining existence (Folke 2006).
Since first proposed in the 1960s, the idea of alternative stable states and the concept of resilience have deeply influenced studies of population ecology and community response to disturbances. Ecologists have been working on data to increase empirical support for the idea that a community can be found in multiple stable states and that disturbances in state variables and/or parameters (depending on the intellectual context) can force communities to shift from one state to another and still maintain their functional characteristics (Holling 1973, Sutherland 1974, May 1977, Dublin et al. 1990, Laycock 1991, Knowlton 1992, Peterson et al. 1998).
The argument for the existence of alternative stable states and its intrinsic characteristics, such as resilience, has expanded under the ecosystem perspective to incorporate anthropogenic activities (Nystrom et al. 2000, Beisner et al. 2003. In social sciences, the resilience perspective has made important contributions. One of the first and most cited uses of the concept of resilience in the realm of social science is attributed to the work of Vayda and McCay (1975), a methodological piece discussing new directions in ecological anthropology. One of the points made by the authors concerns criticisms of ecological anthropology and its preoccupation with static equilibria. Vayda and McCay (1975) focus their argument on human population dynamics and the introduction of the idea of multi-stable systems to understand population fluctuation.
The article by Vayda and McCay (1975) represented an important step for the studies of human populations and the introduction of the idea of multi-stable communities and resilience into human societies (Folke 2006). Since the mid 1970s, trends in social sciences have been shaped primarily by the new ideas introduced by the resilience concept as opposed to an equilibrium-centered view (Klein et al. 2003, Abel et al. 2006, Davidson 2010. The application of resilience theory into social sciences is particularly relevant when addressing societies that are directly dependent on natural resources such as fishing communities. In such societies, social resilience is heavily interconnected with the resilience of the natural system itself, mainly through resource dependency and management actions (Adger 2000). More recently, under a more holistic perspective, resilience has been conceptualized in terms of socialecological systems.
Social-ecological systems (SES) can be defined as complex and dynamic biogeographical units characterized by interactions between human and natural components regulating natural, socio-economic and cultural resources (Redman et al. 2004). The concept of SES is used to emphasize the idea that humans and nature are interrelated and that a separation between human social and non-human ecological units is artificial and arbitrary (Folke 2006). The emergence of the SES perspective was part of extensive paradigm shifts that took place over the past few decades in environmental studies and natural resource management, from the idea of simple and linear natural systems where humans are separate agents or 'stressors' to a view of nonlinear complex systems where humans are an integrative component (Kates et al. 2001, Berkes 2004. Over the past one and a half decades, the term SES gained considerable momentum in the literature and a great deal of researchers have focused their work on conceptualizing resilience of SESs (Berkes and Folke 1998, Carpenter et al. 2001, Walker et al. 2002, Adger et al. 2005).
Complex SESs with multiple domains of attraction are characterized by unpredictable responses and shifts in the face of disturbances. Gunderson and Holling (2002), in adapting the resilience framework to understand such complex systems, developed a multiscalar system called panarchy that includes adaptive cycles.
According to panarchy theory, complex systems feature multiple scales formed by interactions between variables performing at similar speeds. The different levels experience their own change cycles but lower, larger scales set the conditions for faster, smaller ones, while faster, smaller cycles produce variations that can, in turn, generate shifts at larger scales. This dynamic interaction among scales fuels adaptation and therefore evolution of complex systems (Davidson 2010).
The aforementioned multi-scale interactions are represented in figure 4 by the panarchy model. The model depicts two dynamic systems each presenting four different phases of the adaptive cycle: a phase of exponential change (r), that is characterized by growth and exploitation and is the preceding stage for a phase of conservation (K), which is characterized by relative stability; a phase of release (Ω), characterized by collapse and severe modifications to the system, which is followed by a phase of renewal (α), which in turn precedes another phase of exponential change (r). The transition between conservation (K) and release (Ω) characterizes regime collapse and can happen rapidly at the peak of system intricacy and productivity. The two systems illustrated in the panarchy model are linked by the elements revolt and remember. The element revolt characterizes disturbance (Ω) of a small and fast cycle on large and slow events, usually breaking a state of conservation (K). The element remember, on the other hand, is related to the learning processes that will help a system in the phase of conservation (K) to cope with the renewal phase (α) in a smaller and faster cycle.
The outcomes of the interaction between different systems will depend on intrinsic characteristics and the size and nature of the perturbations affecting them . Three possible general outcomes exist: the system can reorganize and maintain the same structural regime without changes to structure or function (persistence); the system can shift to a different state, with different feedback processes but within the same regime (adaptation); or it can switch to a new regime with different processes and functions (transformation). Holling et al. (2002) identified three qualities a system must present to avoid transformation: 1) accumulate rather than deplete resources, 2) contain destabilizing forces to maintain diversity and opportunity and stabilizing forces to maintain productivity, and 3) ensure evolutionary processes that generate novelty. In sum, the panarchy theory implies that a complex system must find a balance between dynamism and persistence, diversity and conservation in order to adapt and thrive in the face of disturbances (Davidson 2010).
The adaptive cycle and the panarchy model proposed by Gunderson and Holling (2002) are, by definition, heuristic representations, for the interactions displayed in them are not based on data assessment but on theoretical constructs. The importance of such a representation lies on it being a means of illustrating dynamic systems in which change is an integrative component. In an earlier publication, Holling (1973) emphasizes the intellectual significance of a perspective that steers away from equilibrium centered views and embraces the idea of change: "flowing from [the resilience theory] would not be the presumption of sufficient knowledge, but the recognition of our ignorance; not the assumption that future events are expected but that they will be unexpected" (1973:21).
The application of the resilience concept into SES research has contributed to the expansion of the theory into different disciplines and the incorporation of a vast number of other concepts into the resilience rationale (e.g. social capital, leadership, role of institutions, governance, and climate change). All these contributions to resilience theory have undoubtedly enriched this concept and its potential applications but have also added a great deal of complexity to its conceptualization, often compromising consensus regarding definition and application of the theory (Gallopín 2006). Furthermore, a great deal of debate concerns the relationships between the ideas of resilience and vulnerability. Depending on the conceptual definitions used, disciplinary framework, and personal preference, resilience and vulnerability are used interchangeably, as opposite ends of a continuum, or as concepts complexly interrelated (Cutter et al. 2008, Miller et al. 2010, Turner 2010 It is not among the objectives of the present study to discuss the ramifications of the conceptual framework for resilience theory, but rather to attempt to apply the concept in a practical perspective in the context of U.S. fishing communities. The literature presents a number of different conceptual definitions for resilience (see Walker et al. 2004, Carpenter et al. 2001, Folke 2006, Gibbs 2009, Cutter et al. 2009). The definition adopted in the present study is: the ability of a system to cope with change in the face of specific disturbances that can be of social, political, natural or economical nature. Under the idea of 'coping with change,' a resilient SES undergoing pressure may not necessarily go back to a previous state of equilibrium, but may change to a different desirable state. Based on this idea, managing human-natural systems for resilience involves putting in place mechanisms to maintain diversity and homeostatic functions while steering them away from thresholds of concern and increasing their ability to cope with change (Allen et al. 2011).

Resilience of fishing communities
Negative changes in fisheries are a global concern (Jackson et al. 2001, Worm et al. 2009). Demands for natural resources and the pressures of a growing population have impacted fish stocks while also causing regulations on the exploitation of natural resources to increase and become stricter (Ostrom et al. 1999). In the face of changing environments, declining fish stocks, and increasing regulations, there has been growing concern for the adaptation and resilience of fishing communities (Allison et al. 2007, Coulthard 2012, Robards and Greenberg 2007, Healey 2009). Resilience is the ability of a system to cope and adapt to change.
Therefore, resilient systems are adaptable, flexible and prepared for change and uncertainty. In fishing communities, resilience outcomes refer generally to long-term sustainability of jobs, identity and culture, together with healthy fish stocks and resilient ecosystems , Coulthard 2012.
The resilience concept is often linked to the idea of sustainability (Folke et al. 2004, Maler 2008. The overall idea is that resilient SESs are more capable of achieving sustainable development (Lebel et al. 2006, Perrings 2006. Although the direct relationship between resilience and sustainability in SESs has been contested (see Derissen et al. 2011), in general terms, it can be argued that examining attributes of a system that are more or less resilient to certain disturbances has the potential to favor the adoption of practices of sustainable development that are a better fit for such system as opposed to practices that do not take resilience into account (Turner 2010).
This is especially relevant in fishing communities in the current context of fish stock decline and frequent policy changes.
Increased efforts have been made in the area of coastal hazards to operationalize resilience as well as vulnerability at the community level (Sempier et al. 2010, Cutter et al. 2008), but very little has been done with regard to fishing communities. One of the few efforts made specifically in fishing communities is a study by Jacob et al. (2013), who developed indices of resilience and vulnerability 2 for fishing communities in the Gulf of Mexico. The indicators used by Jacob et al. (2013) were developed with a secondary data baseline and measure resilience and vulnerability at the community level. The authors explain that developing indicators of resilience and vulnerability in fishing communities can be "very useful in a social impact assessment framework for local governments, regional agencies, and national planning" (Jacob et al. 2013). Ongoing efforts by the National Atmospheric and Oceanic Administration (NOAA) are being made to expand the use of these indicators into other regions of the U.S. (Lisa Colburn, personal communication, 2013).  provide another example of an effort to measure resilience in fishing communities. Different from the study by Jacob et al. (2013), the authors used survey questions to assess individual subjective resilience among commercial fishermen in northern Australia.  focused specifically on responses of fishermen to changes in fisheries policy. The authors criticize the implementation of management strategies that fail to consider social impacts on fishing communities and argue that this lack of consideration for social aspects such as resilience, i.e. the ability that fishermen have to cope with policy changes, leads to poor compliance with the rules and regulations . The measures of subjective resilience developed by  are used in the present study and will be examined in Chapter III.
The literature describes resilient system responses in terms of absorption of change, reorganization, self-learning, and innovation (Armitage andPlummer 2010, Folke 2006). Coulthard (2012) describes three possible ways that fishermen could cope and reorganize in the face of changes in the fishing industry: exit strategy, livelihood diversification, and remain fishing. The first two possibilities -leave the fisheries and seek alternative occupation -are confronted by aspects of well-being that are very important among fishermen, especially aspects of job satisfaction. According to Pollnac and Poggie (2008) fishermen frequently describe fishing as more than just an occupation, and numerous examples in the literature show that fishermen would resist leaving the occupation of fishing even when income is low (Crawford 2002, Binkley 1995, Pollnac et al. 2001).
Job satisfaction is an important component related to workers' well-being in any occupation, but it is especially significant in fishing jobs. Among fishermen, the structure of job satisfaction includes attributes of 'adventure,' 'challenge' and 'being outdoors' infrequently found in other occupations (Binkley 1995, Apostle et al. 1985, Pollnac and Poggie 1988. Research in industrialized countries demonstrates that job satisfaction affects individuals' health and productivity and it is related to mental health, longevity and social illnesses such as family violence and substance abuse ).
Levels of job satisfaction have also been associated with personality traits (Bruk-Lee et al. 2009). Pollnac andPoggie (2006, 2008) argue that individuals with a personality type characterized as active, adventurous, aggressive, and courageous are attracted by and attached to activities and professions, such as fishing, that enable fulfillment of these needs. Also, for many fishermen, occupation attachment is developed and reinforced by familial traditions and interactions with other fishermen during and outside of working hours ).
The more attached people are to their occupation, the more difficult it is to either leave it or deal with significant changes to it. According to : When people with a strong occupational attachment suddenly face the prospect that they are no longer able to continue in their current occupation, they not only lose a means of earning an income, they lose an important part of their self-identity. (2007: 364) While examining resilience of fishing communities it is important to consider aspects of cultural resilience, or the capacity to maintain livelihoods that suit both material and moral needs in the face of change and social dynamism to avoid negative cultural transformations (Crane 2010). According to Poggie et al. (1995), "commercial fishing is an occupation that has all the prerequisites for being considered an occupational subculture" (1995: 411). Among the reasons for that, is the fact that fishermen present distinct ideas and behaviors and are subject to factors and pressures that are unique to their life-style (Poggie et al. 1995). The cultural component of the occupation of fishing reinforces the idea that changes to the occupation of fishing have the potential to impact more than just fishermen's source of income.
Considering the attachment that fishermen manifest towards their occupation, it is more likely that they will remain in the fisheries, even after impacts imposed by declining stocks and strict regulations. It has been shown in different parts of the world that programs to reduce fishing pressure based on alternative occupations as well as boat buyback programs fail to achieve foreseen goals because many fishermen will not adhere or they will use the money earned to improve their fishing practices (Pollnac et al. 2001, Crawford 2002, Sievanen et al. 2005. For that reason, strategies that aim at maintaining resilience of systems highly dependent on fishing must consider issues of job satisfaction, identity and occupational culture in order to preserve both community and ecosystem resilience.

Diversity and 'adaptive flexibility' of fishing communities
Diversity is associated with flexibility, variability, redundancy, and adaptability: all attributes believed to characterize a resilient system (Ebbin 2009). At the ecosystem level, resilience can be defined in terms of diversity of habitat, species, and trophic levels (Holling 1995). Deriving from the ecological perspective, diversity has also been associated with governance and institutional resilience (Jones et al. 2013, Adger et al. 2005, Ebbin 2009). In social systems, occupational and income diversity have been regarded as key components to adaptability (McCay 2002, Allison and Ellis 2001, Adger 2000, Carpenter et al. 2001, Cinner et al. 2010, Abesamis et al. 2006). Specifically in rural and fishing communities the importance of diversity has been widely discussed in the literature often under the umbrella of the 'livelihood approach' (Hanazaki et al. 2013, Ha and van Dijk 2013, Marschke and Berkes 2006. The concept of 'livelihood' encompasses assets and activities (mediated by institutions and social relations) through which people can generate a satisfactory standard of living (Krantz 2001). Assets can be natural, physical, human, financial or social capital (Ellis 2000). In the context of fisheries, these assets could be specified respectively as fish stocks and fishing grounds, boats and gear, labor and personal experience, savings and credit, and kinship and social networks (Ha and van Dijk 2013). According to Allison and Ellis (2001), the most favorable livelihood displays "high resilience and low sensitivity" (2001:378). Sensitivity in this context can be described as the degree to which a given system is modified or affected by internal or external disturbances (Gallopín 2006). In the context of fishing communities, the concept of favorable livelihood is closely related to the health and sustainability of the fishing resources. According to Chambers and Conway (1992) a 'sustainable livelihood' can cope and recover from stresses, "maintain or enhance its capabilities, assets and entitlements, while not undermining the resource base" (1992: 6).

Cinner et al. (2010) discuss the importance of diversification to livelihoods in
African rural, fishing communities through the principle of 'portfolio' or the spreading of risk. According to the authors "the adoption of a diverse portfolio [is] expected to contribute to the sustainability of rural livelihood because they improve resilience in the face of seasonality, adverse trends, and sudden shocks" (2010: 23). Portfolio theory focuses on the diversification of assets as a means to ensure the most favorable expected outcome (e.g. catch or income), an idea rooted on patterns of statistical averaging and correlations among assets or portfolio options (Sethi 2010). Statistical averaging theory is based on the idea that the sum of given variables (diversification) has lower variance than the individual variables alone (specialization), resulting in greater stability (Doak et al. 1998). Diversification of assets also takes advantage of correlation divergences among variables (Elton and Gruber 1977). In that sense, the opportunity for exploiting resources with very different characteristics increases the chance of a favorable outcome in the face of adversities.
The logic behind the idea of diversification of assets is expressed in the concept of 'polymorphic adaptation,' widely employed in biological sciences to describe the process of organisms' adaptation through diversity of forms, physiology, and genetics (Johnson et al. 1996, Majerus 1998. In anthropology, Parkin and Ulijaszek (2007)  The idea of livelihood diversification is characterized by an evolutionary approach. A great deal of literature defends the viewpoint that the theory of evolution can be generalized to socioeconomic phenomena (see Dennett 1995, Rammel andvan der Bergh 2010). Under this context, alternative occupations, technologies, and sources of income provide the evolutionary potential to adapt to changes, a phenomenon that can be defined as 'adaptive flexibility.' In biology, adaptive or behavioral flexibility is seen as an important adaptation to changing environments, which, at the individual level, may arise through innovation or social learning (Wright et al. 2010). Rammel and van der Bergh (2010) argue that the evolutionary thinking behind the idea of adaptive flexibility in human societies is a "fruitful approach to study policies for sustainable development" (2010: 122).
According to Adger (2000) " The literature supports the idea that risks associated with a loss of diversity in economic activities are more important in communities that are highly resource and ecosystem dependent (Adger 2000, Kasperski andHolland 2013). Such communities are more likely subject to fluctuations of these resources as well as potentially destructive natural hazards. Fishing, similar to other activities that involve extraction of natural resources, is characterized by the need for effective responsiveness to changes in the external environment (Pitcher 2001). These changes are not only related to the physical environment such as weather and oceanographic features but also to biological fluctuations in fish populations, management decisions influencing availability of different species, as well as market variations. Figure 5 illustrates the relationships between human and natural environments and highlights changes in the fisheries that affect fishing community resilience that are pertinent for the current study. As mentioned previously, fishermen are in general highly attached to their occupation, and there is significant evidence in the literature to believe that they are unlikely to search for alternative sources of income outside the fishing industry (Pollnac et al. 2001, Acheson 1981, 1988, Griffith and Pizzini 2002, Glazier 2007, Ginkel 2007, Smith and Clay 2010. American fishermen have other compelling reasons to persist in their occupation; in developed countries, fishermen are generally more invested in and attain larger incomes from fishing. These factors are likely to increase dependency on fishing (Allison and Ellis 2001), which strengthens the argument for 'within-fishing diversification' as opposed to alternative employment as a strategy to increase fishing community adaptability and ultimately resilience to changes in the fishery.
Different studies conducted in the past couple of decades have shown the importance of 'within-fishing diversification.' While studying small-scale fishermen adaptation to uncertainty, Allison and Ellis (2001) state that one of the strategies implemented by fishermen to deal with fishery resource fluctuations was targeting different species according to availability. Marschke and Berkes (2006) found that in rural Cambodia the ability of fishermen to access and use diverse types of fishing gear, therefore having flexibility to switch gear depending on season and resource abundance, was regarded as an important indicator of well-being among villagers. In a study comparing vulnerability of fishing and farming families in Congo, Bené (2009) found that fishermen who were specialized in one species were more vulnerable than those who targeted a variety of different species.
Recently, growing efforts in the field of risk management have discussed the stabilizing effects of diversity, especially in the context of natural resources management (see Koellner and Schmitz 2006, Sethi 2010, Tilman et al. 2006. Studies on risk management in the U.S. and in other developed countries have stressed the importance of 'within-fisheries diversification' (portfolio theory) in reducing catch variance and providing a buffering mechanism in the face of change (Hilborn et al. 2003, Baelde 2001, Minnegal and Dwyer 2008, Schindler et al. 2010). Kasperski and Holland (2013) investigated income diversification and risk among U.S. West Coast and Alaska fishermen and concluded that vessels that are able to diversify across multiple resources can reduce income variation and the associated financial risk.
According to Sethi (2010), the portfolio theory is applicable as a risk management strategy because it "not only increase[s] efficiency, but also reduce[s] the exposure to both biological and economic variability in fishery systems" (2010: 354).
Policy strategies conventionally used to manage the fisheries in the U.S., however, are generally not aimed at maintaining or inducing fisheries diversity (Whitmarsh 1998, Rammel and van der Bergh 2010, Sethi 2010. In fact, the majority of regulations being implemented to control fishing activities in the U.S. today tend to reduce fishermen's flexibility to diversify. Management can influence fishermen's adaptive flexibility in a variety of different ways. Under rights based limited entry programs, such as catch shares and other quota limitation systems, fishermen harvest species they have quota to catch and avoid those for which they have no quota, thus constraining their options and consequently their adaptive flexibility. As species fluctuate in numbers, holding different licenses is one way to guarantee the fishermen flexibility to continue to fish and make a living (Apollonio and Dykstra 2008). Murray et al. (2010) while studying the effects of "enclosure" as a consequence of the implementation of limited entry programs in New Jersey fisheries concluded that this management system led to "loss of flexibility for fishers who depend on moving among fisheries" (2010: 3). The authors used oral histories to discuss the negative consequences of limited entry to adaptability at the individual and community level as well as the potential environmental impacts of specialization due to intensified pressure on fewer resources.
Limited entry programs can also affect fishermen's adaptive flexibility in different ways. The process most often used for determining qualification for limited access programs is "historical landings" for established qualifying years. This means that vessels that historically rotate between resources have a disadvantage when compared to vessels that fished heavily for fewer stocks. In fact, vessels that have diversified their fishing practices to alleviate pressure on certain resources (often in response to NMFS requests) have been denied licenses to stocks they were helping conserve because they did not qualify due to lack of historic landings (Crocker 2008).
Consequently, the anticipation of being able to guarantee a license for a certain fishery leads fishermen to focus their effort on fewer resources, inhibiting their natural inclination to diversify and putting even more pressure on specific stocks.
The implementation of closed areas for the conservation of specific stocks can also promote specialization. Large closed areas affect access to all resources within a geographical scope and, according to Apollonio and Dykstra (2008): [Reduce] the option that fishermen have for responding to natural variations in the numbers of various species [and] limits their ability to shift from one species to another, as they have traditionally done for generations, as species naturally fluctuate in numbers or distributions. (2008: 78) While fishermen are forced out of different fisheries, the possibility of accessing other resources is decreasing. Under an economic focused regimen, prices of licenses and quotas have increased to the extent that they are unaffordable to many fishermen. Furthermore, fishermen who were able to secure multiple licenses are so constrained by bureaucratic processes that it can be virtually impossible to make an unanticipated shift in target species and/or fishing grounds.
All the above mentioned constraints imposed on the fishermen suggest that even though specialization in the fishery has known negative social, economic, and environmental consequences, management practices in the U.S. are in general not designed to promote diversity. Fishing is subject to continuous changes and impacts: variation in catch due to fish stock fluctuation and seasonality, weather, tides and currents, fishing ground attributes, market fluctuations, and management. The means and ability to exploit an array of different resources at different times of the year constitutes part of the adaptive tools fishermen have to overcome change and guarantee a stable flow of income. In other words, diversity of catch functions to maintain adaptive flexibility and to ensure long-term survival in the face of adversities, a concept that can help shape beneficial strategies to safeguard community and cultural resilience.   (Anthony 1990). Early on in the new management regime, the body of law regulating the fisheries off the coast of New England experienced a considerable increase in complexity. Groundfish 8 fisheries were initially managed under a quota system, which was the structure inherited from ICNAF. However, this management system motivated a "derby" fishery, i.e. a race for fish, and allocations set for the entire fishing year were taken within a few months (Crocker 2008). To address that problem, new quota allocations were implemented on a quarterly basis, and trip limits by species, weekly landing limits and quotas by vessel class were established (Apollonio and Dykstra 2008). Between 1977 and 1981, one hundred and twenty modifications were made to the fishing regulations (Hennessey and Healey 2000).

Fisheries management in New England
Increasing the complexity of rules did little to achieve the goals of the NEFMC of conserving fish stocks. Fishermen were "forced" to throw dead fish overboard to 7 The ICNAF was established in January 1949 during a conference held in Washington DC involving ten other countries. The convention was a consequence of a concern over declining fish stocks throughout the Northwest Atlantic. The ICNAF came into force in July 1950 following ratification by Canada, Iceland, the UK and the USA. The stated purpose of ICNAF was the "investigation, protection, and conservation of the fisheries" (NAFO 2013). 8 Also called Northeast Multispecies, the groundfish fishery comprises 15 species regulated under one FMP. Twelve species are regulated as large mesh species (based on the size and type of gear used): Atlantic cod, haddock, pollock, yellowtail flounder, witch flounder, winter flounder, windowpane flounder, American plaice, Atlantic halibut, redfish, ocean pout, and white hake. Three species are regulated under a small mesh program: silver hake (whiting), red hake, and offshore hake (NEFMC 2013b).
comply with quota limitations and frequently disrespected the regulations imposed. A fisherman as quoted by Crocker (2008) stated about the new quota systems: "Fishermen didn't want to waste fish or break the rules like that, […] but the system created an untenable situation that undermined a generations-old stewardship ethic and brought out the worst in human nature" (2008: 2).
By the early 1980s, stock assessments indicated that three of the major stocks of groundfish -cod, haddock, and yellowtail flounder -had declined (National Research Council 1998). For that reason, the Council discontinued the quota system in favor of indirect controls for regulating the fisheries off the coast of New England (Jin et al. 2002). Regulations under the indirect controls approach aimed at controlling mesh size, fish size, and area closures. Rules were incorporated into two different groundfish management plans, the Interim Groundfish Plan, implemented in 1982, and the Multispecies Plan, implemented in 1986.
According to Healey and Hennessey (1998), the Interim Groundfish Plan of 1982 "incorporated considerable complexity in an attempt to deal equitably with the diversity of New England fishery" (1998: 111). In order to attend to this diversity, the fishery plan defined areas where cod, haddock, and yellowtail flounder could be caught with mesh sizes larger than the minimum requirement and also offered an "optional settlement program" for specific small mesh species, as to not exclude fishermen who fished for species other than cod, haddock, and yellowtail flounder in the large mesh areas. Fishermen participating in the "optional settlement program" had to land 50 percent of their catch consisting of small mesh species, while only 15 percent could be of cod, haddock, and yellowtail flounder, and they were not required to send in logs of their fishing activity. This program allowed fishermen targeting small mesh species to carry nets of different sizes on the boats and change them accordingly to their fishing activity (NEFMC 1981).
Between 1982 and 1984 it was concluded that the Groundfish Interim Plan did little to conserve the fish stocks. Fishermen frequently ignored provisions regarding mesh sizes and complied with fish size rules mainly by discarding and high grading.
Closed areas required intensive monitoring by the Coast Guard to be effective. The complexity of the regulations and the difficulty in enforcing them, as well as the perception by the fishermen that the rules lacked fairness, contributed to the overall failure of the plan (Healy and Hennessey 1998 were not allowed to fish in groundfish fishing areas unless certified by a NMFS Regional Administrator that the vessels could not achieve a groundfish by-catch of more than 5 percent of the trip catch. Amendment 7 also put in place a Multispecies Monitoring Committee responsible for tracking DAS and TAC utilization and making recommendations to the council regarding FMPs (NEFMC 1996).
Part of the justification for increasing regulations concerning inshore fishing grounds was a consequence of the DAS proposed rules in Amendment 5 (Apollonio and Dykstra 2008). Vessels that fished mainly offshore (e.g. Georges Bank), looking for maximizing their DAS fishing season by reducing steaming time, began fishing in inshore waters. This created further pressure on inshore stocks and increased competition among fishermen who traditionally fished in those waters versus newcomers seeking new grounds and species not included in the DAS program. For that reason, NMFS later implemented Amendment 8 in an attempt to minimize gear conflicts as a result of the regulations previously imposed (Apollonio and Dykstra 2008). The council stated: Abundance of stocks fluctuate[s] from natural causes and many fisheries resources are harvested well above the level that would maximize yield. As a result, fishermen have learned to make the most efficient use of their capital and labor by targeting other species, possibly with different fishing gear. This pulse fishing may be economically efficient, especially when an underutilized species serve as a substitute for a depleted species in the marketplace. It may not be so beneficial to the fishery resources or the ecosystem. It certainly concentrates fishing effort in areas where the new targeted species occurs and may disturb fishing activities that already occur in that area. (NEFMC 1996) The concern expressed by the council regarding gear conflicts reflects fishermen's natural inclination to diversify in the face of changes. In the specific case of Amendment 8, changes imposed were in part a consequence of DAS regulations, which caused an imbalance in the distribution of fishermen in the different fishing grounds (Apollonio and Dykstra 2008). In an attempt to resolve conflict, the NEFMC response was to further restrict fishermen's ability to diversify.
The year of 1996 was also marked by the reauthorization and amendment of the MSA with the enactment of the Sustainable Fisheries Act (SFA), which shifted the focus of the MSA from promoting national fisheries to conserving fish stocks. Among other provisions, the SFA required the removal of discretion over the definition of overfishing, (i.e. prevented managers from setting catch limits above sustainable levels to achieve short-term economic gain (PEW 2011)) and the requirement that rebuilding of overfished stocks be done within a specific time period (generally ten years).
Moreover, the act implemented new requirements to reduce by-catch and waste as well as provisions to protect essential habitat for fish populations. The MSA reauthorization of 1996 also put a moratorium on the implementation on any new Individual Fisheries Quota (IFQ) system in U.S. fisheries for five years (Sustainable Fisheries Act of 1996, U.S. P.L. 104-297, 11 October, 1996). In 1997, NFMS released the first "Status of Fisheries of the United States" report to Congress indicating that eighty-six species of fish were considered overfished (Dell'Apa et al. 2012).
In March 1999, Amendment 9 was implemented and it further revised definitions for overfishing and specifications of optimum yield for 12 groundfish species. Amendment 9 also includes Atlantic halibut in the multispecies plan (64 Federal Register 55 13952-13953, 23 March, 1999). Later in 1999, Amendment 12 9 to the MSA addressed the management of the small-mesh groundfish species (silver hake (whiting), red hake, offshore hake, and ocean pout) including a moratorium on the commercial permits for these species (64 Federal Register 104 29257-29258, 1 June, 1999). In June of the same year NOAA announced Disaster Assistance for Northeast Multispecies Fisheries Failure, which included a plan to disburse funds to those who have incurred losses from declining fish stocks (64 Federal Register 112 31542-31548, 11 June, 1999).
In 2000, the Conservation Law Foundation along with four other organizations once again sued NMFS for failing to address the issue of overfishing of cod, haddock, and yellowtail flounder. The years that followed the lawsuit were marked by the 9 Amendment 10 reflected administrative adjustments to the plan and Amendment 11 set Essential Fish Habitat for all species of groundfish (Apollonio and Dykstra 2008 (2002), and the Groundfish Peer Review Panel (2003) (Brodziak et al. 2008). All these efforts culminated in the implementation of Amendment 13 to the New England multispecies FMP in 2004. By the time the amendment was implemented, NMFS was facing five lawsuits on the grounds that it failed to comply with the MSA to protect New England groundfish stocks (Apollonio and Dykstra 2008).
Interestingly, in July 2001, the Stock Assessment Review Committee (SARC) reported to the New England council that the Gulf of Maine cod stock biomass had more than doubled in two years (1999-2001) but could not explain the reason why that happened while fishing mortality was approximately three times the recommended level. The Fmax, or the fishing mortality that ensured the maximum sustainable yield, for the cod stock in question was 0.27 and it was around 0.7 for the period of time assessed in the report. Despite the results presented by the SARC, NMFS advised the NEFMC that rules be set to meet the biological target (Apollonio and Dykstra 2008).
Amendment 13 was implemented with that goal and put in place new rules concerning the DAS program aiming to decrease fishing mortality by approximately 60 percent, and consequently further increase the complexity of the New England groundfish FMP. Among other adjustments to the plan, four new categories of permitted DAS were established (Brodziak et al. 2008). A council member, as quoted by Apollonio and Dykstra (2008), stated: "the current management system needs to be simplified. The combination of time, gear, and area restriction is daunting at best and incomprehensible at worst" (2008: 68). Back in 1999, the following quote was published in Commercial Fisheries News as quoted by Apollonio and Dykstra (2008): The rules have become so complicated and they are changing so often that no one, absolutely no one, can keep track of them all… Even the people at the National Marine Fisheries Service who have to implement the rules admit they can hardly keep up and are having trouble adequately informing fishermen of the latest in the endless stream of changes. (2008: 65) In addition to all the amendments to the groundfish FMP, between 1994 and 2004, more than forty framework adjustments had been implemented. Framework adjustments are effected without the need for the full administrative review required for amendments.
Besides establishing new rules to the DAS system, Amendment 13 also introduced for the first time the idea of "sectors" and set rules for the creation of the Georges Bank Hook Gear Sector Allocation (NEFMC 2003). The new rules to the groundfish management were also responsible for eliminating a large number of fishermen from the fishery by not allocating them any viable DAS, and for reducing the DAS of remaining vessels causing financial hardship to many fishermen (Holland et al. 2010). By 2006 NEFMC recognized the complexity and consequent problems of the DAS system and considered alternative management options such as a "points" system, "area management," and an expanded "sector" program (Apollonio and Dykstra 2008). The council considered implementing an IFQ program in the groundfish fishery but was afraid that it might not pass a referendum vote by two- The primary reason for the implementation of LAPPs, or any catch share system in the fishery, was believed to be to create an incentive for fishermen to fish sustainably (Schikler 2008). Under catch shares fishermen are granted a property right over a portion of the TAC and therefore would feel more compelled to conserve the fish stocks. This reasoning derives from the idea that 'ownership promotes stewardship' (Costello et al. 2008, Stokstad 2008. Another reason stated for implementing a system of limited access to the groundfish fisheries was to end the race for fish. It is argued that giving fishermen the right to a portion of the catch, therefore eliminating the incentive to race other fishermen for fish, increases safety and lengthens fishing seasons. According to Schikler (2008) "if implemented properly, harvesting rights give fishers an enforceable right to exclude others from the fishery and encourages sustainable behavior that will ensure a long-term flow of benefits from these assets" (2008: 914). According to Apollonio and Dykstra (2008), one main argument for the council to vote against the implementation of limited access plans was that once implemented, it would change aspects of the social and economic characteristics of New England fisheries. Furthermore, it was recognized that because of the complicated and interrelated fishery practices of the New England fisheries, implementing a limited access system in one fishery would deeply affect other fisheries in ways that no one could predict (Apollonio and Dykstra 2008).
Despite the controversial nature of limited access programs, in the early 2000s the council was once again debating the implementation of such management strategy in New England fisheries, convinced that a system of days-at-sea was extremely complicated and no longer effective to control effort of a fleet that had grown more Vessels organized into sectors became exempt of the effort controls previously implemented in the fishery, such as DAS limits. Vessel owners who opt for not participating in a sector formed the "common pool," subject to controls of DAS, trip limits, and all area closures. After Amendment 16 passed, seventeen sectors 11 were created and each established rules for allocation use of the total quota granted to the sector as a unit. Allocations were granted based on historical catches in the groundfish fishery from a fixed period (1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006), resulting in 98 percent of the TAC allocated to sectors, while 2 percent was granted to the "common pool" vessels, which represented 46 percent of all vessels in 2010 (Kitts et al. 2011). Under the new market-based approach, sectors and their members can lease or trade allocations from fishermen in other sectors, creating an internal market for fishing privileges (Labaree 2012).
The implementation of sectors in the New England groundfish fisheries under Amendment 16 sealed the transition of the fisheries management scheme in the region from an effort control system to rights based limited access management. From the beginning, fishery participants were split with regard to their support over the implementation of sectors and some of the main criticisms concerned the fear of consolidation, increased difficulty of entry for new participants, and decreased fleet diversity (Macinko and Whitmore 2009). It is true that these issues would exist in the groundfish fishery regardless of the implementation of sectors, but the nature of limited access programs is known to facilitate and even motivate them. In Alaska, for example, the implementation of catch-shares to the red king crab fishery contributed to a decrease of approximately 65 percent in the total number of vessels (Knapp 2008).
Specifically with regard to sectors, permit holders are allowed and even encouraged to pool allocations into fewer vessels (Macinko and Whitmore 2009). It is also argued that the method used for the initial allocations to sector members favored larger vessels, therefore further compromising fleet diversity. A number of other impacts to fishermen and fishing communities have also been considered (Olson 2011, Olson et al. Forthcoming 2013. One example, a survey conducted by the Gulf of Maine Research Institute in 2011, showed that a significant number of groundfish permit holders had dropped out either permanently (selling out) or temporarily (leasing out) because they believed their "potential sector contribution" was not enough to stay profitable (Labaree 2012).
It is undeniable that the implementation of Amendment 16 posed substantial changes to the groundfish fishery in New England and possibly to entire communities.
According to Macinko and Whitmore (2009: 38) "sectors will not provide protection to communities unless such protection is built in." In 2012, the NEFMC considered Amendment 18 to the groundfish FMP, which, if passed, would put in place "measures that will impose limits on the amount of allocations that individuals or groups of individuals may control" and could "create other incentives for maintaining diversity and fishery infrastructure" (NEFMC nd). Amendment 18 is the council's recognition that current management practices have threatened the flexible and diverse characteristics of the New England groundfish fisheries.

Scallop fishery
The Atlantic sea scallop fishery ranges from the Gulf of Maine to the Mid-Atlantic and it represents one of the most valuable fishing resources to the region today and the most valuable wild scallop fishery in the world (NOAA 2010). Although five stock components 12 are recognized by science, they are managed as one single unit by NMFS, with the exception of the Gulf of Maine stock (Repetto 2001). The Atlantic Sea Scallop FMP was first implemented by the NEFMC in 1982. The stated objective of the plan was to "maximize overtime the joint social and economic benefits from the harvesting and use of the sea scallop resource" (NEFMC 1982).
Until 1993, the scallop resources were managed almost exclusively by a "meat count," or the maximum number of scallop "meats" 13 that was contained in one pound of shucked scallops. Entry into the fishery was open until 1994, when the stocks were declared overfished and a moratorium on the issuance of new permits was put in place.
Amendment 4 of 1994 established the rules to the new limited access scallop management system. Approximately 350 fishermen qualified for licenses, representing basically all fishermen who could prove significant harvest history in the previous years (Repetto 2001). Licenses were broken down into full-time, part-time, and occasional, with the majority of granted licenses (264) representing full-time vessels (Edwards 2002). To prevent permit "stacking," licenses could not be disengaged from vessels, and therefore could only be transferred upon sale or transfer of the vessel itself. This type of program is referred to as non-transferable Individual Vessel Effort Quotas (IVEQs). Fishermen who did not qualify for limited access had the option of applying for a "general category" permit, also effected with Amendment 4, and were allowed to land up to 400 pounds of scallops a day (Hartley 2010). Other measures to address overfishing were also adopted. Days at sea were scheduled to decrease from 200 to 120 over the years, crew size was limited to a maximum of seven men to constrain the amount of scallops that could be shucked on a trip, and minimum diameters were established for the rings on dredges to allow small scallops to escape (Repetto 2001).
In December 1994, due to the collapse of the groundfish fishery, three areas of Georges Bank were closed to all vessels that could potentially catch groundfish, including scallop boats. This measure culminated with a shift in effort to other areas in the Mid-Atlantic and small open areas, which caused these areas to be subsequently closed to protect juveniles. The groundfish and scallop closed areas represented about one-third of the scallop resource area and approximately 85 percent of harvestable biomass (Edwards 2002). After the reauthorization of the MSA in 1996, more drastic measures were proposed to further decrease effort in the scallop fishery. In 1998 it was proposed that the allowable days at sea would fall from 120 to only fifty-one (Repetto 2001). These measures generated a response from permit holders, who in 1999 created the Fisheries Survival Fund, a group dedicated to lobby for access to closed areas.
Research funded by the industry revealed that stocks had increased eight to sixteen- The overall intent of this action is to stabilize capacity and prevent overfishing from the general category fishery, to maintain the diverse nature and flexibility within this component of the scallop fleet, and preserve the ability for vessels to participate in the general category fishery at different levels (NEFMC 2007: 1).
Although maintaining diversity and flexibility was a stated objective of Amendment 11, there was growing concern by many fishermen and researchers at the time that the implementation of limited entry programs (in this case an IFQ) in the scallop fishery could have negative social and economic consequences for many fishing communities (see Olson 2006).

Surveys
Structured surveys conducted with fishermen from the ports studied were employed as a means of assessing and measuring their perception of changes in flexibility and diversity, as well as the processes behind these changes. Surveys were also deemed as the most appropriate method for obtaining data on the potential effects that a decrease in diversity can have on fishermen resilience. Moreover, surveys were also used to investigate correlational patterns between fishing related variables and subjective perceptions of resilience and job satisfaction among fishermen.

Participants and sampling design
A total of 117 fishermen were interviewed from the ports of Point Judith, Rhode Island, New Bedford and Fairhaven, Massachusetts, and the Cape Cod region of Massachusetts, more specifically the ports of Chatham, Harwich Port, Hyannis, and Provincetown. For the purposes of this study the studied communities will be referred to as the ports of Point Judith, New Bedford, and Cape Cod.
The method used for sampling individuals was a direct approach at the docks, characterizing an opportunistic sampling technique (Bernard 2006, Rudestam andNewton 2007). The reason for choosing this sampling design was due to the challenges involved in drawing a truly random sample from the universe studied.
There is currently no registry of active fishermen available for the studied area, making a random approach at the docks the most effective way to contact them, which is especially true when attempting to reach crew members. Moreover, fishermen often live far from the ports they sail from, and their schedules are difficult to predict.
Fishing is at the mercy of weather conditions, seasonal fluctuations in fish stocks, and regulations, which will determine when they can sail and for how long they will be away on a trip. It has been determined by previous research (Pollnac and Poggie 1978: 365) that the most successful way to obtain information from fishermen is to approach them at the docks when they are working gear, preparing to leave on a trip, coming back from a trip, or simply socializing with other fishermen. All these activities are difficult to predict, making it further challenging to draw a truly random sample from the universe of fishermen. In an attempt to avoid biases, fishermen were approached at random days of the week and times of the day. A sample obtained in this manner can be conceptualized as a sample from the universe of all hypothetically possible data sets collected under similar conditions (Chein 1976, Freund 1960, Thomas 1976).
Interviews lasted an average of fifteen to twenty minutes.

Measures
The questionnaire (Appendix I) was designed to 1) obtain information about demographics and aspects of respondent's fishing activity; 2) obtain information on fishermen's perceptions of changes in catch composition through open ended questions; 3) assess individual subjective resilience; and 4) assess individual subjective levels of job satisfaction.

Subjective resilience
Levels of subjective resilience were measured at the individual level using a scale developed by . The scale consists of twelve statements designed to assess levels of well-being among fishermen concerning their flexibility, opportunities, and acceptability with regard to changes in the fisheries. The scale replicated in this study was the result of a reliability analysis involving seventeen initial statements developed by Marshal and Marshal (2007). The twelve statements selected (table 4) represented those with a Cronbach's α of 0.7 or greater 14 .
14 Cronbach's α is a type of reliability test used in statistics to calculate reliable generalization to a universe of variables from a sample of variables (Rummel 1970). The test is based on correlations between statements (variables) and it has a maximum value of one. Respondents were asked to indicate their levels of agreement with the statements above and responses were coded on a five-point Likert scale (1 = strongly disagree; 2 = disagree; 3 = neutral; 4 = agree; 5 = strongly agree). Four statements (indicated with an asterisk in table 4) were negatively worded and were therefore coded on a reversed scale (1 = strongly agree; 2 = agree; 3 = neutral; 4 = disagree; 5 = strongly disagree). The original scale developed by Marshal and Marshal (2007) was coded on a four-point Likert scale (1 = strongly agree; 2 = agree; 3 = disagree; 4 = strongly disagree -reversed for negatively worded statements). In the current study it was considered important to include a neutral point in the scale, therefore offering respondents the option to neither agree nor disagree with the statements.

Job satisfaction
Job satisfaction among fishermen was first assessed by Pollnac and Poggie in 1977(Acheson et al. 1980, Pollnac and Poggie 1988 and then by them and many others in the U.S. and Canadian fisheries (e.g., Gatewood and McCay 1990, Pollnac and Poggie 2008, 1988, Binkley 1995, Apostle et al. 1985. This variable was originally measured using a twenty-two-item scale including topics that were shown by previous research to be associated with job satisfaction among fishermen. Factor analysis involving the original scale was used to develop the three components of job satisfaction: Basic Needs, Social and Psychological Needs, and  Table 5. Items derived from the twenty-two item scale developed by Pollnac and Poggie (1988)  and 'opportunity to be own boss' the Self-Actualization component; and 'time away from home,' 'physical fatigue of the job,' and 'healthfulness of the job' the Social and Psychological Needs component (Pollnac and Poggie 2006). Two other job satisfaction questions ("Would you advise a young person to go into fishing?" and "Would you still fish if you had your life to live over?"), previously used by Pollnac and others, were also used as job satisfaction indicators. Responses to these two questions were coded as yes, maybe, or no.

Post-survey supplementary questions
After obtaining results for the interviews conducted using the questionnaire described above, it was regarded appropriate to conduct additional fieldwork to query fishermen more directly concerning perceptions of their flexibility to exploit an array of different species or to use multiple gear types and about potential changes in flexibility through time. Due to the fact that surveys were administered in Cape Cod ports after the decision of obtaining further data, fishermen in these ports were asked the additional questions concurrently with the questionnaire described above.

Participants and sampling design for supplementary questions
A total of forty fishermen were interviewed between the ports of Point Judith and New Bedford using the supplementary questionnaire and twenty-five fishermen in Cape Cod were asked the supplementary questions along with the main questionnaire.
The sampling method used was the same described for the main surveys and fishermen were approached at the docks to comply with the same methods and considerations described previously. In Point Judith and New Bedford, where supplementary questions were asked alone, interviews lasted on average five to ten minutes.

Measures
Post-survey supplemental questionnaires (Appendix II) consisted of 1) a brief demographic section and questions on fishing attributes; 2) questions regarding the level of flexibility to exploit different species and switch gear types both at the time of the interview and when respondent first began fishing; and 3) follow-up questions regarding reasons and times of changes reported.

Flexibility questions
Questions regarding flexibility to exploit different species and to switch gear types were intended to capture potential changes that fishermen may have perceived during their fishing experience with regard to their flexibility. Responses were coded on a Likert scale to allow quantitative analysis of reported changes. The scale used ranged from one to fifteen, with one being the lowest possible level of flexibility -or no flexibility -and fifteen being the highest possible level of flexibility -or absolute flexibility. For both themes regarding flexibility to exploit different species and to use different gear types, a question was asked for the current time of the interview and for when the respondent first began fishing, both coded on the same Likert scale ranging from one to fifteen.

Follow-up questions
Fishermen who reported any changes with regard to their flexibility to either exploit different species or use multiple gears or both were asked follow-up questions regarding perceived reasons and time of the change mentioned. These questions were open-ended questions that were later coded according to categories of responses.

Sample
In order to investigate potential changes in diversity of species caught through time, catch composition of vessels homeported in the studied areas for the period between 1994 and 2012 were analyzed. The landings data was obtained from the National Oceanic and Atmospheric Administration (NOAA) commercial fisheries databases with permission granted. Information on gear type and pounds of species caught by vessel were obtained for New Bedford and Fairhaven (referred to as New Bedford), Narragansett (referred to as Point Judith), Chatham and Provincetown 15 (referred to as Cape Cod) for the eighty-eight different species listed in Appendix III.

Diversity measure
The landings data was used to calculate a measure of diversity that could be assigned to each landing by vessel to represent the level of contribution of different species to the overall catch composition of a particular vessel in a given year. The measurement chosen was the Shannon Index, which has been extensively used in the 15 Due to characteristics of the data obtained through the databases and necessary data transformations it was decided that the two largest ports in the Cape Cod region and those with the most significant number of fishermen surveyed in this study, Chatham and Provincetown, would be used. ecological literature to quantify ecosystem diversity. The Shannon Index is based on the idea that a higher number of species signifies a more proportional abundance, therefore higher diversity. The closer the Shannon value is to zero the less diversity exists in the system. The Shannon Index is calculated by the equation: where: H = Shannon diversity index Pi = proportion of the entire population belonging to species i R = the number of species categories encountered ∑ = the sum of all species The Shannon Index was deemed appropriate for the purposes of this study because values of diversity obtained for each vessel landing can be used to investigate changes in diversity, or proportions of contribution by different species to overall catch composition, through time for the ports studied, using analysis of variance.
Landings data was also used to examine co-occurrence of species landed in order to investigate potential fluctuations in landings by combinations of species through time using principal component analysis and analysis of variance.

Analyses
Results obtained from survey questions as well as from the landings data were analyzed using statistical tests for the appropriate measurement levels on Systat® software.

CHAPTER IV ANALYSES AND RESULTS
This chapter presents analyses conducted with data obtained through the use of surveys presented in the previous chapter as well as analyses conducted using landings data obtained through assessment of NOAA Fisheries databases. This chapter will provide information necessary for discussing the research hypotheses.

Age and education
The total sample of fishermen (n = 117) from the ports of New Bedford (n = 41), Point Judith (n = 51), and Cape Cod (n = 25) had an average age of 46.8 years (SD = 11.740) and 12.6 years of formal education (SD = 2.202). Table 6 presents results of basic statistics for age and education for each port separately and all ports combined.

Marital status
Overall, more than half of the sample was married (61%) and only one person chose not to answer the question. In New Bedford and Point Judith the percentages of married fishermen were just above half (58 and 57%, respectively) and in Cape Cod married fishermen represented 72%. Differences observed were, however, not statistically significant (χ 2 (2) = 1.73, p>0.05). The overall rate of divorced fishermen was relatively low (11%), and differences between ports did not show statistically significant results (χ 2 (2) = 2.92, p>0.05). The majority of married fishermen (76%) stated that their spouses also had an occupation. The lowest incidence of spouses with an occupation was observed in New Bedford (60%). Point Judith and Cape Cod presented similar results (83 and 87% respectively). Differences between ports with regard to the frequency of spouses with an occupation were statistically significant (χ 2 (2) = 6.24, p<0.05).

Residency
The majority of fishermen in the total sample (65%) lived in the same state where their boats were homeported but in a different town. The frequency of fishermen living out of the state was relatively low (8%). The port with the highest incidence of out of state fishermen was New Bedford (17%) and the port with the highest incidence of fishermen living in their homeport town was Cape Cod (42%).
Point Judith presented the highest incidence of fishermen who lived in their homeport state but in a different town (82%). The differences observed were statistically significant (χ 2 (4) = 16.65, p<0.01). Tables and figures with frequencies with regard to fishermen's residency can be seen in Appendix IV.

Occupations besides fishing
The majority of fishermen in the sample (79%) did not have any additional occupations besides fishing. New Bedford and Point Judith presented a higher percentage of fishermen without additional occupations (80% and 84% respectively) when compared to Cape Cod (68%). These differences, however, were not statistically significant (χ 2 (2) = 2.77, p>0.05). A list of all occupations mentioned by fishermen interviewed and their respective frequencies can be seen in Appendix V.  Table 7 shows the results for years of fishing experience overall, on respective homeports, and on current boats for each port studied and all ports combined.

Familial involvement in fishing
The overall average of generations involved in fishing was 2.1 (SD = 1.343).  Table 8 presents the results of basic statistics for number of generations involved in fishing for each port and all ports combined. Out of the total sample of fishermen interviewed, about half (55%) reported the presence of relatives who are also involved in fishing and just below half of them (45%) reported the presence of relatives fishing on the same boat as them. A comparison between ports did not show statistically significant results (χ 2 (2) = 1.99, p>0.05). Table 9 presents results of basic statistics for the number of relatives involved in fishing for each port studied and all ports combined.

Fishery position
Overall the position with the highest incidence in the sample (n = 45) was 'captain/owner' (also known as 'owner/operator'), followed by 'captain' (n = 30), 'crew' (n = 29), 'mate' (n = 10), and finally 'owner' (n = 3). In New Bedford the fishery position most frequently found was 'captain' (n = 18), followed by 'crew' (n = 10). In Point Judith and Cape Cod the most frequently observed fishery position was 'captain/owner' (n = 21 and n = 18, respectively), followed by crew (n = 15 and n = 4, respectively). Table 10 shows the distributions of the different positions across the three ports studied and figure 6 provides a visual representation of the distribution found.   Table 11 shows results of basic statistics for crew size for each port studied and all ports combined. Cod and New Bedford were the ports with the shortest (M = .86, SD = .55) and longest (M = 7.9, SD = 3.87) trips respectively. Table 12 shows basic statistics for trip length for each port studied and all ports combined.  Figure 7 shows a visual representation of the differences between ports with regard to frequencies of trip length reported by fishermen interviewed. While Cape Cod shows a concentration of responses that characterize a majority of short trips, Point Judith shows a more diversified range of trip lengths with the majority lying between short and medium trips, and New Bedford shows a bimodal distribution of trip lengths with some short trips and a majority of long trips.

Permit type
The majority of fishermen interviewed (91%) fished on boats with federal permits. The frequency of boats with federal permits was similar across all ports (χ 2 (2) = 0.25, p>0.05). Table 13 shows frequencies for presence of federal permits on boats in the overall sample and in each port studied.  Figure 8 shows a visual representation of the distribution of income categories by port.   Figure 8. Histogram comparing the distribution of income categories across the three ports studied

Gear types
Overall, the most common primary gear type in the sample was trawl (dragger) (42%) followed by dredges (21%). Other primary gear types found in the overall sample can be seen in table 16. In New Bedford, the most common primary gear type found was dredge (46%) followed by trawl (34%). In Point Judith the most common primary gear type found was trawl (67%) followed by lobster traps (25%). In Cape Cod the most common primary gear type was gillnet (32%) followed by rod & reel (24%). All primary gear types and their frequencies for New Bedford, Point Judith, and Cape Cod can be seen in tables 17, 18, and 19 respectively.

Sector affiliation
Out of the total fishermen asked if their boats belonged to a sector (n = 98 16 ) about half of them (47%) had an affiliation with a fishery sector, and half (50%) did 16 Total sample size is different because this question was not asked during the first ten interviews in Point Judith and there are seven missing cases for that port and one missing case for Cape Cod regarding this specific question.
not have an affiliation with a fishery sector (3% did not know). In New Bedford and Cape Cod the majority of fishermen interviewed did not belong to a sector (66 and 58% respectively). In Point Judith the majority of fishermen who responded to this question (67%) belonged to a sector. The differences observed were statistically significant (χ 2 (2) = 11.24, p<0.01).

Species targeted
The overall average number of primary species targeted mentioned by the The species most frequently mentioned overall as primary targets, in terms of volume, were scallops (23%) and squid (17%). Although the response with the highest frequency with regard to the existence of secondary principal target species was 'none' (17%), the two species most frequently mentioned as secondary targets, in terms of volume, were fluke and monkfish (both with 10%). In New Bedford, the most frequently targeted primary species was scallops (44%). The second most important primary target species in New Bedford was flounder (10%). The most frequent answer regarding secondary species targeted in New Bedford was "none" (24%). Monkfish was the species with the highest frequency of response (22%) among the fishermen who targeted a secondary species. In Point Judith the most frequently mentioned primary target species was squid (37%) followed by scup (15%). Fluke (24%) and scup (18%) were the species most frequently mentioned as secondary target species in terms of volume. Only one fisherman in Point Judith did not have a secondary target species. In Cape Cod the primary target species most frequently mentioned were scallops (20%) and dogfish (16%). The most frequent response among Cape Cod fishermen with regard to the existence of a secondary target species was 'none' (28%) and the most common secondary target species mentioned was tuna (24%). Tables with all species targeted and their frequencies by ports can be seen in Appendix VII.

Job satisfaction variables
A factor analysis with varimax rotation using the nine job satisfaction questions derived from the scales developed by Pollnac and Poggie (1988) and presented in Chapter III was used to reduce the data into three components. Careful examination of the scree plot and a cut-off Eigenvalue of one were used as criteria for selecting the number of components. The results reflect groupings of variables that are similar to those found by Pollnac andPoggie (1988, 2006). The components Basic Needs, Social and Psychological Needs, and Self-Actualization can be identified using the highest factor scores for each variable, with the exception of the variable "opportunity to be your own boss," which was found along with variables characterized by the component Social and Psychological Needs as opposed to the component Self-Actualization where it had been previously grouped. Table 27 presents the results for the factor analysis of job satisfaction variables. Factor loadings above 0.40 were considered significant.

Job satisfaction and gear types
Levels of job satisfaction were also compared among different primary gear types. In order to accommodate the diversity of gears found in the sample and to avoid small sample sizes by gear type, gears were grouped into the three categories multispecies, traps, and shellfish gear. The multispecies category encompasses the gear types: trawl, gillnet, and rod & reel; the category traps includes lobster and fish traps; and the category shellfish groups the gear types dredges and hydraulic dredges. Table 28 shows the distribution of the transformed gear categories for each port.

Additional job satisfaction questions
As part of the job satisfaction measure, fishermen were asked about their willingness to become a fisherman if they had their lives to live over and whether they would advise a young person to become a fisherman. Overall, the majority of responses (60%) with regard to advising a young person to enter the occupation of fishing was negative (30% positive and 10% 'maybe' or 'depends') (χ 2 (2) = 43.74, p<0.001). When asked about their willingness to become a fisherman if they had their lives to live over the majority of responses (69%) was positive (25% negative and 6% 'maybe' or 'depends') (χ 2 (2) = 74.05, p<0.001). Chi square analyses comparing the three ports studied with regard to negative versus positive responses were not statistically significant for either one of the job satisfaction questions. The same was found when primary gear type categories were compared. Tables showing frequencies of responses for each port and primary gear type can be seen in Appendix VIII.

Resilience variables
The twelve resilience variables developed by  and presented in the previous chapter were reduced into three distinct components using factor analysis with varimax rotation. Careful examination of the scree plot and a cut-off Eigenvalue of one were used as criteria for selecting the number of components. These results differ from the findings by , in which the variables were reduced into four components. The three components were named Ability to Cope, Perception of Risk, and Adaptation to Change according to the characteristics of the variables with the highest factor loadings for each component.
Factor loadings above 0.40 were considered significant. Table 29 shows the factor loadings for all resilience variables and highlights the highest loadings for each variable that contribute significantly to the three components listed. Although the number of components differs, the groupings of the resilience variables found in the present study are similar to the results found by . The only differences are with regard to the variables "I can cope with small changes in the industry," which in    .

Resilience and ports
The three ports studied were compared on their levels of resilience for each of the three components developed. All ports presented similar mean values with regard to their levels of resilience for each of the components and none of the results for analyses of variance were statistically significant. Table 30 shows mean standardized values for the resilience components across the three ports.  Comparisons involving the other resilience components and multiple gear type users were not statistically significant.

Correlations between variables
Pearson Correlation analyses were used to understand relationships between the various independent and dependent variables. A Principal Components analysis with varimax rotation including demographic and fishing related variables was conducted previously in order to reduce these variables into components that were  In order to investigate the relationships between the components created above and variables of job satisfaction and resilience in the studied sample, a Pearson Correlation analysis was used. Statistically significant correlations were found between the job satisfaction component Social and Psychological Needs and the components Fishing Occupation Attributes, Gear Use and Education, and the job satisfaction variables willingness to advise a young person to enter fishing and willingness to be a fisherman if had life to live over. The component Basic Needs was statistically significantly correlated with the resilience components Ability to Cope and Perception of Risk, as well as with the job satisfaction variable willingness to advise a young person to enter fishing. The resilience components Ability to Cope and Adaptation to Change were statistically significantly correlated with the job satisfaction variables willingness to advise a young person to enter fishing and willingness to be a fisherman if had life to live over respectively. A correlation between the two job satisfaction variables willingness to advise a young person to enter fishing and willingness to be a fisherman if had life to live over was also observed. Table 33 shows the correlation matrix between all the different components and highlights the statistically significant relationships. Figure 9 shows a heuristic model based on the statistically significant correlations between the different components.  (1) = 24.2, p<0.001). Analyses of the two ports separately showed the same patterns observed for the overall sample. In New Bedford (n = 41), 76% of respondents believed catch composition to have changed (χ 2 (1) = 10.76, p = 0.001) and in Point Judith (n = 39 19 ) fishermen with the same opinion accounted for 79% (χ 2 (1) = 13.56, p<0.001). Chi square analyses comparing the two ports did not show statistically significant results.
Fishermen were queried on the time they believed changes in catch composition to have begun to occur as well as reasons for changes mentioned. Overall, fishermen believed changes to have begun around 1998 (SD = 15.02). In order to investigate potential patterns of response by different age groups with regard to the time changes were believed to have begun to occur, the sample was dichotomized using the mean age for New Bedford and Point Judith combined as a cut-off point (M = 46, SD = 10.12). Fishermen younger than 46 years believed changes to have occurred around 1999 (SD = 21.4) and fishermen older than 46 years believed changes to have occurred around 1998 (SD = 9.33). A comparison between the two age groups with regard to the average year changes have occurred was not statistically significant (t(23) = -.39, p>0.05, separate variance).
Overall, the majority of fishermen interviewed (73%) believed regulations to be the main reason for changes in catch composition. Other answers involved stock depletion or overfishing (11%), market fluctuations (10%), and climate change (2%) 20 .

Description of the sample
A total of sixty-four fishermen from Point Judith (n = 30), New Bedford (n=10), and Cape Cod (n = 24 21 ) were asked supplementary questions aimed at investigating their opinion with regard to changes to their level of flexibility, in other words, changes to the ability of diversifying their fishing activity in terms of target species and gear type. The average age of the sample was 50 years (SD = 11.5) and the average years of education was 12.5 (SD = 2.02). The average fishing experience was 29.9 years (SD = 10.32) with 24.6 years of experience in the current port (SD = 12.0). Table 34 shows results of basic statistics for age, education, and fishing experience for the overall sample. The fishery position with the highest frequency in the sample was 'captain/owner' (n = 28 (44%)), followed by 'captain' (n = 16 (25%)). Table 35 shows the frequencies for all fishery positions and figure 10 shows a visual representation of the distribution. 21 One missing case   Table 36 shows the distributions of all gear types found in the overall sample. Out of the sixty-one fishermen who were queried about their affiliation with a fishery sector, a statistically significant majority of 66% responded affirmatively (χ 2 (1) = 5.92, p<0.05).
The fishermen interviewed using the supplementary questionnaire listed a total of twenty-nine target species. The species with the highest frequency among responses was squid (n = 28), followed by fluke (n = 19), scup (n = 16) and groundfish in general (n = 14). A list of all species listed and their respective frequencies can be seen in Appendix IX.

Flexibility questions
Responses given by fishermen interviewed using the supplementary questionnaire with regard to their levels of flexibility today and when they first began fishing concerning the diversity of species they can exploit and the types of gear they can use were coded on a Likert scale ranging from one to fifteen. The two time periods were paralleled using paired sample t tests to compare the point in the scale where fishermen consider themselves to be today and where they were when they first began fishing. Results for differences between the two time periods were statistically significant both with regard to diversity of species and gear types, and in both cases mean levels of flexibility were perceived to be considerably higher when fishermen began to fish in comparison to the present time. Results of the paired sample t tests comparing levels of perceived flexibility to exploit different species and to use different gear types are shown in tables 37 and 38, respectively.  Fishermen were asked about the time period in which the perceived changes mentioned, if any, began to occur. The overall mean time period was 1993 (SD = 10.83). In order to test differences between generations with regard to the perceived mean time period of changes, the sample was dichotomized using the mean age (M = 50) as the cut-off point. Fishermen who were younger than 50 years (n = 19) perceived changes to have occurred in the year 2000 on average (SD = 8.5) and fishermen who were older than 50 years old (n = 27) perceived changes to have occurred in 1988 on average (SD = 9.93). Changes with regard to the flexibility to exploit different species and use multiple gear types maintained the same statistically significant patterns observed for the entire sample when tested for the two age groups separately (results can be seen in Appendix X).
Fishermen who mentioned changes with regard to their flexibility to exploit different species or use multiple gear types were also asked their opinion about the reasons behind the changes mentioned. A total of fifty-two fishermen in the sample responded to this question and 90% of them believed regulations to be the main reason for changes in flexibility. Other reasons mentioned more than once were overfishing or too many fishermen (n = 5 (9.6%)), pressure from environmental groups (n = 3 (5.8%)), and flawed science (n = 3 (5.8%)). Table 39 shows all the different reasons believed by fishermen interviewed to have affected their flexibility to exploit different species and/or use multiple gear types. Total amount of responses does not equal total amount of fishermen who responded to the question (n = 52) and total percentage does not equal one hundred percent because some respondents stated multiple reasons. The maximum amount of reasons per respondent was three and only one person stated three different reasons. The gear types most commonly found in New Bedford were trawl, dredge (scallops), dredge (other), and pots/traps (other). In Point Judith the gear types with the highest incidence were trawl, pots/traps (other), pots/traps (lobster inshore), and handline. In Cape Cod the most common gear types found were longline, gillnet, handline, and trawl. Table 40 shows the total number of landings reported for each of the main gear types for all three ports. followed by a general decrease and then maintaining low fluctuation patterns for the last four or five years. The same trend was observed in a by port basis. Figure 11 shows the chart resulting from the analysis of variance involving all three ports for number of gear types reported by vessel across time. Figure 11. Results of an analysis of variance involving the three ports analyzed comparing the average number of gear types reported by vessels across the time period studied Number of gears by vessel

Diversity analyses
The landings data was analyzed to investigate changes in diversity of species landed at the vessel level using the Shannon Index. Landings for a total of eighty-eight species were analyzed between 1994 and 2012 (a list of all species can be seen in Appendix III). Overall, results show a decrease in the Shannon value, indicating a decrease in diversity in the landings between 1994 and 2012 in terms of the contribution of species to the total weight landed by each vessel/landing (F(18, 6,401) = 10.2, p<0.001) (figure 12).

Species landings fluctuation
The species included in the diversity analyses above were analyzed for fluctuations in landings through time. The total number of species was reduced by first

Cape Cod
The group composed by the species red fish, pollock, and Atlantic halibut showed some fluctuations throughout the years but landings presented a general tendency to increase between 2004 and 2008 for all three ports ( figure 15), more noticeably so in New Bedford (F(18, 3,282) = 5.95, p<0.001).   Bedford (F(18, 3,282)

Primary findings
The principal findings of this study will be discussed in this section. Overall, the findings provide interesting new insights on issues involving the impacts of fisheries management on catch composition and fisheries diversity in the New England region. The results also provide meaningful information about the social structure of fishing communities and how this structure may have or can be affected by change. An additional, brief characterization of the ports studied will be provided based on the results obtained with the analyses of fishing related variables, and major findings of this research will be discussed in the context of the two hypotheses developed and presented in Chapter I.

Results of analyses involving fishing attribute variables showed that New
Bedford, Point Judith, and Cape Cod are considerably distinct from each other with regard to characteristics of their fishing activities, a fact that is believed to enrich the findings of the present research. New Bedford is mainly characterized by the presence of larger vessels, generally sailing on long offshore trips of about a week with large crews of generally five and up to seven people. The involvement of large corporations in the port, due primarily to the lucrative scallop business, is evidenced by the lower occurrence of captain/owners in the sample and a larger number of captains, most likely working on vessels belonging to "shore owners." The prevalence of scallop fishing also makes New Bedford the port with the highest average annual income among the ports studied. This is likely to attract fishermen from farther areas, and, combined with the large scale of the New Bedford fisheries, possibly explains why the port has the highest incidence of out-of-state fishermen when compared to the other ports studied. The predominance of scallop fishing in New Bedford also makes it the most specialized port, with the least number of different species being targeted by vessels.
Point Judith, although a relatively large port, was characterized by a majority of medium sized trawler vessels, making relatively short trips lasting approximately three days on average, and crew sizes rarely exceeding five people. Point Judith presented a relatively diverse fishery, with a primary focus on pelagic species such as squid and scup. The relatively high incidence of captain/owners in the sample likely indicates lower corporative influence in the port, especially in contrast with New Bedford.
The Cape Cod region was characterized by a majority of smaller boats, in general day-trippers, and with small crews typically not exceeding three people.
Although the most common primary gear type found in the sample was gillnet, the most frequently mentioned target species was scallop. Other important resources were dogfish and tuna. These results emphasize the relative diversity of the Cape Cod fishery regarding both target species and gear types. The high incidence of captain/owners and relatively small crews in Cape Cod are indicative of small business operations. The high incidence of fishermen living in the same town as their homeports likely indicates that these businesses are mainly controlled by local people.

Hypothesis I
The first hypothesis developed in this study states that diversity in New England fisheries, specifically in the ports and regions studied, has decreased as a consequence of fisheries management practices, more so for fishermen using On the other hand, results of analyses involving the landings data show that, although diversity in catch composition has decreased to a certain degree, they are not nearly as alarming as the survey data suggests. The analyses of variance comparing Shannon Index scores throughout nearly two decades show that diversity in catch composition has decreased significantly only in New Bedford and in Cape Cod, the latter only when trawl gear landings were analyzed separately.
The higher levels of specialization observed in New Bedford can be in part explained by the success of scallop fishing in the port. It is possible that fishermen consciously became more specialized as the market value for scallops increased considerably in the past decades. However, New Bedford is also one of the ports in the New England region with the highest incidence of groundfish permits, and when landings by trawl gear were analyzed exclusively, levels of diversity also showed a slight overall decrease in New Bedford, which supports the first hypothesis that diversity has declined in the fishery for reasons other than vessels turning to scalloping because of its high value. The fact that Cape Cod landings also show a decrease in diversity when trawl landings were analyzed also helps to reinforce the idea that specialization is occurring in the fishery to a certain degree. The overall results involving the landings data, however, especially when Point Judith is considered, do not entirely support the findings of the surveys.
One possible explanation for the relatively high level of diversity observed in the analyses of landings data is that, although fishermen have been dealing with substantial changes to their occupation due to regulations and have been experiencing restrictions with regard to the flexibility in their decision making, they were still able to diversify and possibly did so in response to these changes. As was extensively The general frustration expressed by fishermen may not yet be observable in landings diversity patterns, but that does not mean they will not occur in the future.
The data shows that fishermen were able to diversify to a certain degree and cope with significant changes from 1994 through 2012, but their strong belief that their flexibility to exploit different species and use different gear types has been compromised by regulations can be seen as an indication that regulations which negatively affect diversity are likely to hinder fishermen's ability to cope with change in the future.

Hypothesis II
The second hypothesis developed states that a reduction in flexibility and diversity in the fishery has the potential for negatively affecting individual and community resilience. This hypothesis proved to be challenging to support because If it keeps the way it is, soon the waterfront will be gone. We support a lot of people. (Captain, personal communication, New Bedford, November 2012) Regulations are believed to impact fishermen's adaptability in a variety of ways. Some fishermen reported that they have failed to qualify for quota for certain species that they could target and benefit from in the future, therefore decreasing their flexibility, because they were targeting different species during specified qualifying years, sometimes for conservation reasons: I was fishing for monkfish when they did the groundfish allocations -staying away from groundfish. I didn't get any groundfish (Captain/owner, personal communication, New Bedford, November 2012) Government changed groundfishing days, took availability away. Now we're fishing for squid. If we could go groundfishing we'd be making more money now. (Crew, personal communication, Point Judith, July 2013) The possibility that fishermen may not qualify for quota for certain species poses a threat to their future adaptation. Nowadays, when available, permits for certain species represent a substantial investment. Moreover, a great deal of fishermen reported that they already depend on leasing or buying quota from other fishermen in order to land a catch that yields favorable revenue. These constraints to adaptive flexibility pose a threat to fishermen's resilience and potentially to the resilience of the resource itself. As it was previously mentioned in this study, the same argument that exists in favor of in-fisheries diversification with the goal of 'spreading the risk' can be applied to the impacts of fishing on the fish populations. The more specialized the fishing fleet are, the more pressure they are likely to put on those specific resources.

Other potentially important findings
Results from this study yielded other noteworthy findings not directly related to the hypotheses developed but that provide information on social aspects of New England fisheries that could potentially be useful in the policy making process, especially in the context of social impact assessments. Results regarding the job satisfaction variables as well as patterns of correlation found among different demographic and fishing related variables and variables of job satisfaction and resilience constitute some of the most interesting outcomes.

Job Satisfaction
The nine job satisfaction variables used in this study have long been applied in fisheries social sciences studies since Pollnac and Poggie (1988) first proposed their use and stressed the importance of the concept when considering adaptation and change in fishing communities (see Bavinck et al. 2012). The three components of job satisfaction (Basic Needs, Social and Psychological Needs, and Self-Actualization) have remained relatively stable in analyses involving the job satisfaction variables in developed and developing countries throughout the decades (Binkley 1995, Pollnac et al. 2001 All these findings strengthen the argument extensively discussed in the literature (Apostle et al. 1985, Gatewood and McCay 1990, Binkley 1995, Pollnac and Poggie 2006) regarding fishermen's prevailing high levels of occupational attachment, especially considering that satisfaction with regard to Self-Actualization, a component deemed as having a particularly important role in fishermen's job satisfaction, maintain high levels across different ports and fishery types. These results suggest that, due to their high levels of job satisfaction, especially with regard to the Self-Actualization component, fishermen in general would likely be reluctant to leave the occupation of fishing despite adversity, an idea that has been previously discussed in the literature (Binkley 1995, Pollnac et al. 2001, Crawford 2002 and is supported by the findings of this study. As discussed in Chapter II, this would decrease their resilience in the face of change.

Correlations between variables
The job satisfaction component as well as to potential changes that could occur in the future. These results suggest that fishermen who perceive their ability to adapt to changes with confidence are more likely to positively evaluate the idea of being a fisherman if they had their lives to live over.
In part, the results obtained with the correlation analyses suggest that monetary gains from fishing play a very important role in fishermen's perception of their ability to adapt and cope with change. One important way that a decrease in diversity in the fishery is likely to impact fishermen directly is by negatively affecting their incomes. Kasperski and Holland (2013) showed that a decrease in diversity in the U.S. west coast fisheries led to a substantial decrease in vessel revenue. The authors described these impacts in the context of loss of flexibility and resilience in the fishery. The correlations found between satisfaction with income and subjective levels of resilience suggest that a decrease in diversity that affects revenue negatively could result in lower levels of perceived ability to adapt to change, i.e. a decrease in resilience among fishermen. England fishermen but it also indicates that flexibility has been compromised to a certain extent. Fishermen expressed a strong belief that their ability to diversify target species and gear types have been reduced as a consequence of regulations, especially due to permit and quota requirements, which has and could diminish adaptability in future scenarios. These findings indicate that impacts on fishery diversity should receive much more consideration during the fishery management process.
The links observed between aspects of job satisfaction among fishermen and levels of perceived resilience, combined with previous findings that show a decrease in levels of job satisfaction among fishermen throughout time in two of the ports studied, suggests that levels of subjective resilience, and therefore fishermen's perception of their adaptability, could have decreased as well. Impacts on well-being and overall perception of the ability to adapt can have an impact on fishermen's willingness to comply and cooperate in the management process ; therefore it is in the interest of policy-makers to maximize resilience and well-being among fishermen.

Conclusions
This study was designed with the main objective of responding the following research question: "have management practices reduced diversity in New England fisheries thus negatively affecting individual and community resilience?" In an effort to answer this question, data on landings occurring between 1994 and 2012 by vessels homeported in three main ports and fishing regions in New England -New Bedford, Point Judith, and Cape Cod -were analyzed for catch composition diversity using the Shannon Index and compared with data obtained through face-to-face surveys with 157 fishermen from the same ports and regions.
Results of analyses show that fishermen perceived a significant reduction in fishery diversity to have occurred as a result of impacts caused by regulations in the past couple of decades, despite the fact that analyses involving landings data showed a significant decrease in diversity only for New Bedford and a slight decreasing trend for New Bedford and Cape Cod when trawl gear landings were analyzed separately.
The latter results can be in part explained by the analyses of landings fluctuation for some of the most important species in the studied region, suggesting that diversification has occurred, possibly as a mean of adaptation to change, which possibly contributed to maintaining stable levels of diversity in landings. It is undeniable, however, that fishermen interviewed strongly believe their flexibility to have been compromised.
This apparent disparity between results involving the two main types of data analyzed in this research raises interesting questions. The present study was not designed to understand, in depth, the rationale behind fishermen's perception of diversity loss, but rather to capture whether or not they believed their flexibility to diversify had changed throughout their fishing experience and the general reason for the change. It is possible that this limitation of the research design could have led to an oversimplification of fishermen's perception of change. However, it is also possible that fishermen expressed a concern for potential future impacts in the face of drastic changes to current important fish stocks, considering that they have lost flexibility to diversify due to regulations -something that would not yet be observable in landings trends. In any case, the results obtained provide promising opportunities for further investigations regarding New England's fishermen's perceptions of changes in diversity and changes to catch composition through time. One significant contribution would be assessing in more depth the rationale behind fishermen's perception of changes in flexibility to diversify and understanding their projections for the future.
Results regarding impacts that changes in fishery diversity can have on fishermen and fishing community resilience, although suggestive of a positive relationship, were not entirely conclusive. Statements by fishermen interviewed showed a relationship between diversity and adaptability, which is not clearly observed in the analyses of the data. Partially, these results can be attributed to the lack of robust measurements for individual and community resilience to changes in the fishery. The measurement chosen in this study -the scale developed by  -constitutes one of the very few attempts at operationalizing resilience at the individual level in the fishery context. It is possible that the measure employed failed to capture levels of resilience associated with the specific context of the present study. It is also possible that the inability to control for other factors that influence subjective levels of adaptability to change distorted the correlation between diversity and subjective resilience.
The correlation between diversity in the fishery and resilience of fishermen and fishing communities, although symptomatic, remains essentially anecdotal and hypothetical. There is great potential for further analyses of the relationship between changes to diversity in the fishery and fishermen's ability to cope. Correlations between fishermen's subjective levels of resilience and changes in fishery diversity could be further evidenced in studies involving a larger geographical scope and comparisons between larger samples of different fishery types to capture in more detail different possible effects of diversity on the adaptability of fishermen. Future research examining changes through time in subjective resilience among fishermen that could be compared with changes in the fishery would be beneficial to help understand fluctuations in perception of levels of adaptability in the future.
Other potential future investigations that would help clarify the issues studied involve the relationship between fluctuations in catch composition diversity and variables of economic nature such as market value for fish and total revenue from landings. These investigations would help understand factors influencing diversity and adaptability in the fisheries. Another possible way to advance the understanding of the relationships between fishery diversity and resilience in the fishery would be to investigate the effects of specialization on fish populations. In addition to evidence found in the literature, results of this study also support the need for further research on the subject.

APPENDIX VII
Primary species landed (volume) for all ports combined: