xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 22 Ka 4500
controlfield tag 007 cr-bnu---uuuuu
008 s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003205
Mercury perception, community awareness and sustainability implications for the tampa bay region, florida
h [electronic resource] /
by Trina Halfhide.
[Tampa, Fla] :
b University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains X pages.
Thesis (M.S.)--University of South Florida, 2009.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
ABSTRACT: Over one million acres of land and water in Florida has been classified as impaired by mercury. Approximately 80% of national fish advisories are issued due to mercury contamination. There have been a number of consumption advisories in the Tampa Bay Region for locally eaten fish such as largemouth bass (Micropterus salmoides), bowfin (Amia calva), and alligator gar (Lepisosteus osseus). The main purpose of this study was to determine if there is adequate dissemination of mercury related risk information by government agencies to recreational and subsistence fishermen in the Tampa Bay Region. This research revealed that government agencies utilized simplified models when addressing mercury consumption risks in Tampa Bay. Most of the popular fishing sites and public parks in the Tampa Bay Region have no advisory signs warning fishers of possible mercury contamination in fish. The majority of survey respondents (88.4%) consumed the fish they caught. There was statistically significant evidence suggesting online sources of public health information influenced viewing of fish advisories. This study determined factors: sex of licensee, above median levels of income and type of license also influenced viewing of fish advisories. Results indicated that women were less likely to view fish advisories than men. In addition, the viewing of fish advisory information by women of reproductive age was not significantly different to all other female age groups. Behavior among participants varied and was dependent on individual perception of mercury risks and nutritional benefits associated with consumption of fish.
Advisor: Fenda A. Akiwumi, Ph.D.
t USF Electronic Theses and Dissertations.
Mercury Perception, Community Awareness an d Sustainability Implications for the Tampa Bay Region, Florida by Trina Halfhide A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Environmental Science and Policy Program Department of Geography College of Arts and Sciences University of South Florida Major Professor: Fenda A. Akiwumi, Ph.D Steven Reader, Ph. D. Amy Stuart, Ph.D. Maya Trotz, Ph.D. Date of Approval: October 29, 2009 Keywords: risk attenuation, risk amplification, fish adviso ries, recreational fishermen, information equity Copyright 2009, Trina Halfhide
Dedication To my dearest brother, Joel Halfhide who forever touches my heart and soul. Keeping busy, I will see you soon.
Acknowledgements I would like to thank my advisor, Dr. Fenda A. Akiwumi and my advisory committee, Dr.Maya Trotz, Dr. Amy Stuart and Dr. Steven Reader for their invaluable guidance and academic contributions to this study. I would also like to thank Dr. Jayajit Chakraborty and Dr. Philip Reeder, whose limited interacti ons have renewed an immeasurable passion for research. I am extremely grateful to have all my d earest graduate collea gues, Joniqua Howard, Richard Salkowe, Nicole Ceasar, Angela Gilbert, Milena Janiec, Emily Zupo, Alec Foster and Marilyn Montgomery, who I call my exte nded family. I will cherish your friendship forever and believe our fina l tears will be ones of joy. I am eternally grateful for my mother, Y vonne Halfhide, whose strength has been an inspiration and a key ingredient in my succe ss. Your love has pushed me through difficult times in this journey. I am extremely grat eful to have a cous in, Laura Young whose prayers strengthened and renewed me. I would like to express my gratitude to Jason Garel, my best friend who provided boundl ess social and em otional support.
i Table of Contents List of Tables ...................................................................................................................... v List of Figures ............................................................................................................... .... vi Abstract ...................................................................................................................... ..... viii Chapter 1: Introduction ...................................................................................................... 1 1.1 Problem Statement .................................................................................... 4 Chapter 2: Literature Review ............................................................................................. 5 2.1 Introduction ............................................................................................... 5 2.2 The fate of mercury in the environment and health endangerment .......... 5 2.3 The role of fish advisories ....................................................................... 10 2.3.1 Responsibilities of agen cies on mercury issues .......................... 11 2.3.2 Legislative restrictions ................................................................ 12 2.3.3 Local and state consumption guides ........................................... 13 2.4 Introductory concepts of sustainability ................................................... 16 Chapter 3: Theoretical Framework ................................................................................... 19 3.1 Introduction ............................................................................................. 19 3.2 Theoretical risk ....................................................................................... 19 3.3 Perceived risks ........................................................................................ 23 3.4 Sustainability as it relates to ri sk perception of mercury in fish ............. 28 Chapter 4: Research Design ............................................................................................. 32
ii 4.1 Introduction ............................................................................................. 32 4.2 Research purpose .................................................................................... 32 4.2 Research questions .................................................................................. 33 Chapter 5: Study Area ....................................................................................................... 34 5.1 Introduction ............................................................................................. 34 5.2 Site selection ........................................................................................... 34 5.3 Physical context ...................................................................................... 35 5.4 Social context .......................................................................................... 38 5.5 Summary ................................................................................................. 39 Chapter 6: Methodology ................................................................................................... 40 6.1 Introduction ............................................................................................. 40 6.2 Introduction to survey sampling method ................................................ 40 6.3 Development of the survey ..................................................................... 41 6.4 Data collection ........................................................................................ 41 6.4.1 Online data collection ................................................................. 42 6.4.2 Data utilized in bina ry logistic regression .................................. 42 6.4.3 In-field data collection ................................................................ 43 6.5 Analyses ............................................................................................................ 44 6.5.1 Online survey Analysis ............................................................... 44 6.5.2 In-field survey Analysis .............................................................. 45 6.6 Key informant semi-structured interviews ........................................................ 45 Chapter 7: Results ............................................................................................................ ..46 7.1 Introduction ............................................................................................. 46 7.2 Online survey results ............................................................................... 46 7.3 Field surveys, direct and participants observations ................................ 50 7.3.1 Field socio-demographics ........................................................... 50 7.3.2 Fishing behavior am ong field participants .................................. 52
iii 7.3.3 Self-reported risk behavior among fishing participants .............. 57 7.3.4 Comparing online and field socio-demographics and risk understanding results ................................................................... 60 7.4 Qualitative results derived from inte rviews with agency officials .................... 64 Chapter 8: Discussion ....................................................................................................... 6 8 8.1 Introduction ............................................................................................. 68 8.2 Research question 1: Are merc ury consumption risks associated with fish being communicated effectively by environmental policy makers (state and local government)? .......................................... 68 8.3 Research question 2: Do all fi sherfolk have equal access to, and understanding of risk consumption information that allows for informed decision making? ..................................................................... 72 8.3.1 Internet media sources ................................................................ 73 8.3.2 Type of license ............................................................................ 74 8.3.3 Sex............................................................................................... 75 8.3.4 Income ......................................................................................... 76 8.4 Research question 3: Does mercury risk information change behavior among fisherfolk? ......................................................................77 8.5 Limitations and future scope for the study ............................................. 80 8.6 Policy implications.................................................................................. 82 Chapter 9: Concluding Statements.................................................................................... 86 References .................................................................................................................... ......87 Appendices .................................................................................................................... .... 95 Appendix 1: Glossary of terms ....................................................................... 96 Appendix 2: List of geocoding databases created ........................................... 98 Appendix 3: Email invita tion to licensees .................................................... 100
iv Appendix 4: Semi-structured survey used in this study ................................ 101 Appendix 5: Semi-structured questions and topics ....................................... 103 Appendix 6: Exemptions to purchasing a license ......................................... 104
v List of Tables Table 5.1: Area of Counties in the defi ned study area (Florida Geographic Data Library data).................................................................................................... 36 Table 5.2: Proportion of area deemed as impaired .......................................................... 38 Table 5.3: Educational attainment of the population census for the respective counties ........................................................................................................... 39 Table 6.1: Summary table with definitions and categories used in binary logistic regression analysis of online survey data .........................................................43 Table 7.1: Descriptive statisti cs for variables analyzed ................................................... 47 Table 7.2: Logit coefficients and odds ra tios from multivariate analysis of self reported viewing of mercury related consumption risk information .............. 48 Table 7.3: Comparison of socio-demogr aphics of particip ants and license purchases ..........................................................................................................48 Table 7.4: Summary table contrasting type of license and level of income ....................50 Table 7.5: Summary table based on se lf-reported demographic questions ......................50 Table 7.6: Summary table for online and field results ..................................................... 61
vi List of Figures Figure 2.1: Regions of Hillsborough cl assified by Environmental Protection Agency (EPA) as being impaired by mercury .................................................. 6 Figure 2.2: Largest 10 national contributor s to global mercury emissions (UNEP, 2008), pp.14 ...................................................................................................... 7 Figure 2.3: Fate of mercury in the envi ronment (University of Wisconsin, Eau Claire, 2009) ..................................................................................................... 8 Figure 2.4: Biomagnification of methyl mercury in the food web hierarchy, adapted using results from (Clec kner 1998) and (Howard, et al. in press, 2009) ....................................................................................................... 9 Figure 2.5: Hillsborough County Environm ental Protection Commission (EPC) consumption guide, (2008) ..............................................................................14 Figure 2.6: Florida Department of H ealth (FDOH) consumption guide, (2008), page 9. Red highlighted region s hows the suggested guidelines for Hillsborough River, Hillsborough County .......................................................15 Figure 2.7: Florida Department of Agricu lture and Consumer Services (FDACS) Swordfish recipes, (2007), page 2. Florida Department of Agriculture recommends eating Swordfish, shark or Tuna twice a week for a good heart................................................................................................................. 16
vii Figure 2.8: The pillars or realms of sustainability. (The World Conservation Union, Report of the IUCN Renowne d Thinkers Meeting, 2006, pg 2) ..........17 Figure 3.1: Risk perception of 81 hazards investigated (Slovic, 1987), pp. 5 .................. 26 Figure 3.2: Graphical representation of carrying capacity and overshoot concept ............21 Figure 5.1: Study sites showing convenien ce sampling and online random survey resident area .................................................................................................... 35 Figure 5.2: Areas being clas sified by Florida Depart ment of Environmental Protection (FDEP) as being impaired ............................................................. 37 Figure 6.1: Survey design and analysis schematic for survey methodology .................... 40 Figure 7.1: Sphere of household influe nce including household age dynamics ................52 Figure 7.2: Rationale for fishing ...................................................................................... 53 Figure 7.3: Fishing frequenc y among participants .......................................................... 53 Figure 7.4: Responses from question 12: Do you eat the fish caught? ............................ 54 Figure 7.5: Fish identified as being commonly caught and eaten by respondents. This estimation was based on a cumulati ve count of the first three fish identified as being eaten by each respondent .................................................. 55
viii Figure 7.6: (A) Fish caught, kept and are st ored in ice troughs. (B) The fish given to occupants upon disembarkation are given their fish for future consumption. All the fish in this bunch were grunt, Haemulon plumier .........57 Figure 7.7: Risk understanding and reported behavioral changes .....................................58 Figure7.8: Verbatim quotes used to analyze risk behavior .............................................. 63 Figure 7.9: Correlation betw een total fish length for largemouth bass (at Loxahatchee) and total log 10 me rcury levels, United States Geological Survey, 2003 ................................................................................. 66 Figure7.10: Mercury poisoning rate (numbe r of persons diagnosed with mercury poisoning per 100,000 persons) using Florida Department of Health 2008 data ..........................................................................................................67 Figure 8.1: Mediated risk Communication Model, adapted from Durant and Lindsey (1999,pp.5) ........................................................................................ 85
ix Mercury Perception, Community Awareness and Sustainability Implications for the Tampa Bay Region, Florida Trina Halfhide Abstract Over one million acres of land and water in Fl orida has been classified as impaired by mercury. Approximately 80% of national fish advisories are issu ed due to mercury contamination. There have been a number of consumption advisories in the Tampa Bay Region for locally eaten fish such as largemouth bass ( Micropterus salmoides ), bowfin ( Amia calva ), and alligator gar ( Lepisosteus osseus). The main purpose of this study was to determine if there is adequate dissemination of mercury related risk information by government agencies to recreational and subsistence fishermen in the Tampa Bay Region. This research revealed that government agencies utilized simplified models when addressing mercury consumption risks in Tampa Bay. Most of the popular fishing sites and public parks in the Tampa Bay Region have no advisory signs warning fishers of possible mercury contamination in fish. Th e majority of survey respondents (88.4%) consumed the fish they caught. There was statistically significant evidence suggesting online sources of public health information in fluenced viewing of fish advisories. This study determined factors: sex of licensee, a bove median levels of income and type of license also influenced viewing of fish advi sories. Results indicated that women were less likely to view fish advisories than men. In addition, the viewing of fish advisory information by women of reproductive age was not significantly different to all other female age groups. Behavior among participan ts varied and was de pendent on individual perception of mercury risks and nutritional benefits associated with consumption of fish.
1 Chapter 1: Introduction Risk perception is an integral component of environmental justice, which is the concept of social stratification based on race, gender, class and pow er distribution ( apek, 1993). Many notable researchers such as Slovic (2008), Kasperson (1988) and Cutter (1993) believe that ones understanding of risks plays a crucial role in behavior and ultimately the risks to which an individual is exposed. Ho wever, in most cases, risk behavior does not necessarily produce a response which w ould reduce a theoretical risk (Kasperson, 1988; Sjberg, 2004). Risk behavior is a function of the perceived risks and benefits of an activity to an individual. The actual beha vioral response of an individual may not mimic the expected response given the theore tical risk. Risk communication by agencies, that are responsible for protecting public health is imperative in reducing health risks to Tampa Bay subsistence and recreational fisher men from the indiscriminate consumption of locally contaminated fish. Maintenan ce of a healthy human population could be considered a success indicat or of sustainability. There are numerous definitions for envir onmental justice; however, this study will employ the United States Environmental Pr otection Agencys (USEPA) definition of environmental justice as it relates to informa tion access. This definition was chosen as the USEPA was one of several agencies respons ible for ensuring no me mbers of a population have disproportionate environmental health burdens. The Clinton Executive Order signed in 1994 outlines that federal agencies have a responsibility in ensuring that programs, policies and activities do not em ploy inclusion or exclusion f actors that may negatively affect or benefit certain raci al or ethnic groups more than others. Environmental justice has been defined by the USEPA (2009) as:
2 The fair treatment and meaningful involvement of all people regardless of race, color, national origin, or income with respect to the development, implementation, and en forcement of environmental laws, regulations, and policies. EPA has this goal for all communities and persons across this Natio n. It will be achieved wh en everyone enjoys the same degree of protection from en vironmental and health hazards and equal access to the decision-making process to have a healthy environment in which to live, learn, and work. It should be noted that although this definition supports the concept that everyone is entitled to equal access to the decision-maki ng process. Many studies have noted that equal access does not necessitate equal outco mes. Unequal access to risk information could hamper the decision making process. Fo r any given person to have equal access to the decision-making process, one must be know ledgeable of the libert y rights and judicial laws which entitle a person to such privil eges. Tampa Bay fishermen and fisherwomen may be considered increasingly vulnerable if they are not aware of local fish advisories. One objective of this study is to determine if there is equal access to information related to mercury consumption risk from recreationa lly obtained fish in the Tampa Bay Region. Most inland populations should not face grave consumption risks from commercial fish. Estimates by the U.S. Department of Health and Human Services, ( 1999) stated that the general United States population consumes commercial fish which commonly contains less than 0.2 ppm methyl merc ury, presenting minimal risk in terms of concentration exposure for those eating a portion size of 32 ounces per week. In addition, few persons eat more than the suggested weekly limit of 35.2 ounces (U.S Department of Health and Human Services, 1999). However, in coastal re gions along the Gulf of Mexico, such as Tampa Bay, population consumption of locally caught fish tends to be higher (Ache et al ., 2000). Most commonly consumed freshw ater fish, such as largemouth bass ( Micropterus salmoides ), bowfin ( Amia calva ), and gar ( Lepisosteus osseus ) as well as 62
3 saltwater fish including King Mackerel and all shark species exceed the USEPA mercury screening level of 0.4 ppm and consumpti on should be limited to the recommended guidelines (Karouna-Renier et al 2008). Florida has the long est coastal border in the United States, which gives Florida resident s easier access to coastal fishing resources than most other states (Karouna-Renier et al 2008). Tampa Bay is a hub for recreational fishing and has seen an incredible growth of recreational boat fishing. It was estimated that there were approximately 125,000 recr eational boats registered in 2005 in Hillsborough, Pinellas, Sarasota and Manatee Counties (Sidman et al 2005). All areas were within 40 miles of the Tampa Bay regi onal coast. Recreati onal and subsistence fishers with extensive access to fishing resour ces, such as those in Tampa Bay, are known to consume larger amounts of fish than the general population (U .S. Department of Health and Human Services, 1999). People who consume greater than 3.53 ounces of fish every day are considered high-end consumer s (U.S Department of Health and Human Services, 1999). High-end consumers eat 10 time s more fish than members of the general population. Subsistence and recreational fisher men routinely fish familiar and similar water bodies, which may increase an individuals exposure to methyl mercury if these waters are contaminated (U.S. Department of Health and Human Services, 1999). Regular fish consumption plays a crucial ro le in maintaining cardiovascular health; however, fish consumption in the United Stat es is the primary s ource of exposure to methyl mercury (Anderson et al ., 2004). Cellular necrosis or death of nerve and muscular cells are expressions of mercury poisoning at the cellular level. Diseases, such as coronary heart disease and Minamata dis ease associated with the consumption of mercury contaminated fish are expressions of mercury poisoning at the systematic level (Counter and Buchanan, 2004; Cernichiari et al 2007). Coastal populations such as Tampa Bay could be considered vulnerable to mercury poisoning as the Florida Peninsula has one of the highest wet deposition levels of total mercury in United States (KarounaRenier et al 2008). Consumption of locally contaminated fish in the Tampa Bay region is a particular concern to hi gh risk groups, such as pre gnant women and children. These vulnerable groups may indiscriminately consume contaminated fish and be subject to the
4 associated health risks. Merc ury exposure from contaminated fish presents insidious risks to the recreational and su bsistence fishing population of Tampa Bay, hence risk communication is imperative in maintaining a healthy coastal community that can make informed health decisions. This study will estimate consumption patterns among participants within th e Tampa Bay Region (Hills borough, Pinellas and Manatee Counties) and determine if recreational fi shing licensees had previously viewed a consumption guide mentioning that mercury is found in high ri sk fish species, such as shark and king mackerel. A semi-structure d survey was conducted between November 2008 and August 2009 at five highly visited publicly accessed fishing areas. These surveys were conducted to determine if ri sks associated with consuming mercury contaminated fish are being understood and i ndividual risk behavior was modified to reduce conceptual risk. Ultimately, this study examined if risks associated with fish are communicated effectively by state and local agencies to all fisherfolk in Tampa Bay. The research was designed to determine if access to risk consumption information allows for informed health decisions. Information access and risk understanding are now being regarded by many researchers and agencies as a necessity to sustain public health in the wake of rapid industrialization (Shr ivastava, 1995; Kar et al 2001). 1.1 Problem statement Humans are intuitive toxicologists, relying on th eir five senses to detect risks, including the risks that are associated with unsafe food (Slovic, 2000) However, mercury does not have a detectable smell or ta ste, and its most toxic form, methyl mercury, is chemically bound to protein and fat in fish. No mercury advisories were posted at the investigated state parks along Hillsborough River or publicly accessed fishing areas in Tampa Bay. This suggests that persons fishing along the Hillsborough River and offGandy Bridge may have limited sources of information at fishing sites. Pamphlets with Hillsborough fish advisory information could be obtained from local doctors offices and agencies
5 headquarters. Most agencies advisories were available online. It is imperative that persons consuming fish, notably recreational and subsistence fishermen are informed and understand the risks associated with the consum ption of fish from contaminated sources. Ignorance of mercury contamination in Florid a fish creates a signifi cant health risk for those individuals who consume this resource. Effective risk communication by agencies relating health concerns associated with c onsuming locally contaminated fish in the Tampa Bay Region is a poorly addressed issue.
5 Chapter 2: Literature Review 2.1 Introduction This chapter is divided into four sections and comprises the following: 1) The fate of mercury and health endangerment, 2) The ro le of fish advisories, 3) Introductory concepts of sustainability and 4) Sustainability as it relates to risk percep tion of mercury in fish. 2.2 The fate of mercury and health endangerment The Southeastern region of the United States, particularly the states along the Gulf Coast experiences some of the countrys highest deposition of mercury (Karouna-Renier et al. 2008). For example, Florida Department of Environmental Protection (FDEP) estimates over one million acres of the Everglades system alone has been impacted by mercury. In addition, the Florida Panhandle, which incl udes the greater Tampa Bay Region, receives some of the highest levels of wet deposition of total mercury in the region. Moreover, the USEPA (2008) identified sites in the Lower, Middle, Upper, and Old Tampa Bay in the Tampa Bay Watershed as impaired due to me rcury contamination. These sites as well as other impaired water bodies are depicted in figure 2.1
Figure 2.1: Regions of Hillsborough classi fied by Environmental Protection Agency (EPA) as being impaired by mercury Mercury is a naturally occurring metal, however most mercury present in the environment can be attributed to anthropoge nic activities (USEPA, 2008). Both local and long-range (non-point) sources contribute to mercury deposition problems in the Tampa Bay Region. Local atmospheric sources include coal-fired utilities, municipal incinerators and industries. Estimates suggest that presen t coal consumption has more than tripled 1930s amounts. Coal-fired power plants are the largest local contributor to Florida mercury emissions (Husar and Husar, 2002). Many researchers argue (Bergan et al ., 1999; Jeffe, et al ., 2005) that the majority of glob al anthropogenic mercury can be attributed to Asian source s and not local sources. Many researchers argue (National Atmospheric Deposition Program/ Mercury Deposition Network, 2009; Karouna-Renier et al 2008) that Florida is a sink for regional and global mercury. Transboundary sources include mining metals and smelting. Mercury is used in many countries including 6
Guyana in the amalgamation process to increase gold yields. Figure 2.2 shows the largest 10 national users contributions to the global emissions. Figure 2.2: Largest 10 national contributors to global mercur y emissions (UNEP, 2008), pp.14 Mercury exists in three oxidation states; Hgo, Hg+ and Hg2+ and can also form organic compounds, with methyl mercury, being the most toxic. Mercury, like other heavy metals such as arsenic and lead tend to accumulate in the lithosphe re (earths sediments) and biosphere (living orga nisms). Mercury undergoes complex interactions from air-water and, watersediment interactions in which biotic and abiotic f actors play a crucial role in its transformation. 7
Figure 2.3: Fate of mercury in the environment (University of Wisconsin, Eau Claire, 2009) When methyl mercury is associated with the sediment, sediment feeders such as Channel Catfish (Ictalurus punctatus) and Quillback ( Carpiodes cyprinus) consume sediment containing mercury. Mercury then bio-accumulate s in an individual fish. Quillback is not a popularly consumed fish among the human po pulation, due to its small size. However, this fish is commonly eaten by game fish such as largemouth bass (Micropterus salmoides) and gar ( Lepisosteus osseus). Methyl mercury and total mercury alike, biomagnifies up the food web with fish some times having 5 times more methyl mercury than periphyton (refer to figure 2.3) (C leckner 1998). Fish feeding on copepods assimilated more methyl mercury than inor ganic mercury due to the larger amounts of methyl mercury present in the soft tiss ues of the copepods (Lawson and Mason 1998). 8
Largemouth Bass Bluefish Red Sunfish Amphipods Water 1E-6 ppm Periphyton 0.10 ppm 3E-3 ppm 3.5 E-3 ppm 0.56 ppm 0.17 ppm Figure 2.4: Biomagnification of methyl mercur y in the food web hierarchy, adapted using results from (Cleckner 1998) and (Howard, et al in press 2009). Most commonly consumed freshw ater fish in Florida include game species; largemouth bass ( Micropterus salmoides ), bowfin ( Amia calva ), and gar (Lepisosteus osseus ) and 62 saltwater fish including King Mackerel and all shark species exceeded the USEPA mercury screening level of ppm (Karouna-Renier et al 2008). Kannan and Faladysz (1998) determined that Hardhead Catfis h and Gafftopsail Catfish in Tampa Bay contained mean methyl mercury of 1.7 and 2.24E-3 ppm (parts per million) wet weight respectively. Largemouth bass from the Hi llsborough River, Tamp a Bay contained 0.56 ppm mean wet weight of total mercury (Howard, et al ., in press 2008). These biochemical and magnification processe s are crucial in influencing how much mercury an individual is possibly exposed to when consuming a given species of fish. Approximately 80% of fish consumption advi sories are due to mercury contamination (USEPA, 2008).When considering the largem outh bass from the Hillsborough River, Tampa Bay which contained 0.56 ppm mean wet weight of total mercury (Howard et al ., 9
10 in press 2009), it is not considered safe to consume by USEPA consumption guidelines. The determination of theoretical consump tion risks are discussed in section 3.2. 2.3 The Role of Fish Advisories For the State of Florida, mercury contaminati on has been a persistent and visible problem for decades, as levels found in fish and other wi ldlife are some of the highest measured in the United States (USEPA, 2008). In 2003, the United States Environmental Protection Agency (USEPA) identified the Hillsborough River as being impaired and a high risk priority area for mercury contamination. Stat e and Federal agencies, such as the Florida Department of Health (FDOH) and the USEP A have posed safety concerns regarding specific fish and shellfish consumption from the Hillsborough River and Tampa Bay. Such concerns have been seen in advisory warnings. For example, advisory warnings for the Hillsborough River can be found on federa l, state and local agencies websites (Florida Department of Health, 2008), (Florida Fish and Wildlife Conservation Commission, 2008). Most advisories specify th e fish or wildlife species of concern; recommended daily or weekly consump tion quantity, fish catch size, and the consumption portion size (U.S. Department of Health and Human Services, 1999). Florida Department of Health (FDOH) established tolerance levels for the consumption of fish and regards fish which have more than 1.5 ppm of total mercury unsafe for consumption. Although fish with concentratio ns between 0.5 and 1.5 ppm total mercury are considered edible, consumption should be li mited to the stipulated guidelines (Florida Department of Environmental Protection, 2008). Fish mercury concentrations with less than 0.5 ppm are considered safe for unlimited consumption for a 156 lb person consuming a single 8 oz. serving of fish per week (FDEP, 2006). Du ring a fish advisory, FDEP recommends limiting the consumption of certain fish in pregnant women and children to once a month, whereas once a week in all other individuals (FDEP, 2006) and may deem some fish such as largemouth ba ss and gar unsuitable for consumption in areas such as Black Creek Canal (C-1). These values were based on a body weight of 70.7
11 kilograms and an 8 ounce (half-pound) serving of fish. It should be noted that the Department of Health issues advisories when the levels of mercury are of public health concern. Fish advisories are not mandated by regulation and are done voluntarily to protect public health (Florida Fish and Wildlife Commission, 2008). 2.3.1 Responsibilities of ag encies on mercury issues In this study five agencies are mentione d: Florida Department of Health, Florida Department of Environmental Protection, Florida Department of Agriculture and Consumer Services, Florida Fish an d Wildlife Conservation Commission and Environmental Protection Commi ssion. The first four are state agencies, while the latter is a local entity of Hillsborough County. All state and local agencies have different statutes, objectives and memorandums which th ey follow. Each agency has its specific role in studying mercury contamination and advising the public how best to reduce their exposure. The agencies role can be summarized as follows: Florida Fish and Wildlife Conservation Co mmission (FFWCC) determines which fish to sample and then collects those resp ective samples. In addition FFWCC conducts laboratory analyses to determine levels within the identi fied water bodies. The Florida Department of Healths role is to determine the potential for adverse human health effects from consuming the local fish and issues fish consumption advisories. Department of Agriculture and Consumer Se rvices provides input on issues related to commercially related seafood in Florida. Th ere are two main interest groups: agribusiness industry and consumers. The Environmental Protection Commission, impl ements federal programs at the local level and issues advisories.
12 2.3.2 Legislative restrictions The Florida Fish and Wildlife Conservati on Commission (FFWCC) outlines the catch restrictions in the recreati onal fishing permits. These regulations are from the FFWWC freshwater fishing regulations: 5 Black bass (largemouth, Suwannee, spotted, and shoal bass, individually or in total), only one of which may be 22 inches or longer in total length. In south Florida. Only one bass may be 14 inches in total length or longer. South and east of the Suwannee River, black bass less than 14 inches in total length mu st be released immediately. 50 Panfish including blue gill, redear sunfish (shellcracker), flier, longear sunfish, mud sunfish, shadow bass, spotted sunfish (stumpknocker), warmouth and redbreast sunfish, indivi dually or in total. 25 Crappie (speckled perch). 2 Butterfly peacock bass, only one of which may be 17 inches or longer in total length (Florida Fish and Wildlife Conservation Commission, verbatim Section1:12, 2008). It should be noted that the above quote outli ned maximum and minimum size restrictions. The maximum size and number of fish capture d were outlined to pr otect human health, whereas the minimum size restriction such as, black bass less than 14 inches in the Suwannee River have to be released were emplaced to protect the fishery resource. Many of the saltwater restrictions outline a minimum size restriction for the game fish such as snapper, to ensure the fishery resource is not being depleted at a faster rate than can be replaced by natural recruitment.
13 2.3.3 Local and state consumption guides Figure 2.5 shows the Hillsborough County E nvironmental Protection Commission (EPC) consumption guide. The guide is intended for children and women of child bearing age. The guide outlines the geographical freshwater area and the level of concern with respect to the species of fish. For example, Largemouth bass ( Micropterus salmoides) is of moderate consumption risk if captured from th e Alafia River, but is of high risk in the Hillsborough River. In addition, the EPC cons umption guide outlines the saltwater fish within the different categorie s of mercury related risk (l ow, moderate and high). Figure 2.6 shows the Florida Department of Health (FDOH) consumption guide and highlights two groups: 1) children and women of child bearing age and 2) other members of the population. It should be noted that in a ll cases FDOH suggests a lower consumption frequency for children and women of child bearing age. FDOH guide like EPCs guide outlined these risks by geographic location for both saltwater and freshwater fish. Although, the FDOH does suggest a frequency consumption for persons that are not within the prescribed high ri sks groups: children and women of child bearing age, neither of the consumption guides outlined men with a history of coronary he art disease as a high risk group. Figure 2.7 shows the Florida Depa rtment of Agriculture and Consumer Services (FDACS) recommendations to eat swor dfish and tuna twice a week to maintain a healthy heart. This recommendation conflicts with both Florida De partment of Health and Environmental Protecti on Commission guidelines.
Figure 2.5: Hillsborough County Environm ental Protection Commission (EPC) consumption guide, (2008). 14
Figure 2.6: Florida Department of Health (FDOH) consumption guide, (2008), page 9. Red highlighted region shows the suggested guidelines for Hillsborough River, Hillsborough County. 15
Figure 2.7: Florida Department of Agricu lture and Consumer Services (FDACS) Swordfish recipes, (2007), pa ge 2. Florida Department of Agriculture recommends eating Swordfish, shark or Tuna twice a week for a good heart. 2.4 Introductory concepts of sustainability The term sustainability has been utilized by many different groups with varying interests including envi ronmentalists, planners, non-govern mental organizations such as the United Nations Development Program (UND P), lawyers, sovereign states such as Massachusetts and businesses. Many have described sustainability as being a relatively new discipline that integrates industrial, social, and envir onmental processes in a global context (Mihelcic et al ., 2003). To gain a firm understanding of the meaning of this term and how it relates to this study, we should ex amine the historical context in which the term was developed and the context in which this term will be used. 16
The sustainability concept was the focal point for the Conference on the Human Environment in Stockholm in 1972, which recognized the dichotomy between maintaining economic growth and environm ental health. The Brundtland Report (1987) definition is the one of the most comm only referenced definitions. Sustainable development was defined in the Brundtland Re port (1987), as development that meets the needs of the present, without compromising the ability of future generations to meet their own needs. pg. 1-2 Sustainable developm ent is prized on being built with three conceptual realms or pillars; the envir onment, the economic growth and society. Society Economic growth Area of sustainability: all three areas are equally considered and valued Environment Figure 2.8: The pillars or realms of sust ainability. (The World Conservation Union, Report of the IUCN Renowned Thinkers Meeting, 2006, pg 2) Many authors and organizations such as the United Nations Environmental Program emphasize that sustainability only exists wh en all three pillars have been equally considered and valued. In most societies, economic growth has a higher priority than the other two pillars: society a nd the environment. Economic growth or development ensures that there is raw material security and w ealth generation. However when the utilization and optimization of the realms are not balan ced, there are long term adverse outcomes. 17
18 Mihelcic et al. (2003) noted efficient use of raw materials may still result in the environmental capacity of an area being surp assed or having social repercussions. Sustainability in this study will be defined as the utilization of natural and human resources in such a manner that does not lead to a diminished quality of life, loss in the economic feasibility of industries or advers e impacts to social conditions, environment and human health. Research and education are essential in pursuing su stainability. One of the main objectives of this study is to increase community awareness of mercury risk associated with locally caught fish and provide information that will enable participants of the study to make informed decisions.
19 Chapter 3: Theoretical Framework 3.1 Introduction Risk can be subdivided into two sections : theoretical risk and perceived risk. A theoretical risk denotes the conceptual probability of a specific eventuality, such as cellular necrosis due to the c onsumption of contaminated fish. Furthermore, theoretical risk is determined by rigorous scientific experimentation and deduction, whilst perceived risk refers to the way in which a risk is interpreted and understood by average persons, not in the risk assessment arena. Theoreti cal risk is usually determined by conducting a risk assessment. 3.2 Theoretical risk Risk assessment is a useful tool for estima ting the likelihood and severity of risks to human health, safety, and the environment as well as making informed decisions in managing those risks. Moreover, it is used to evaluate the potential for adverse health effects from exposure to naturally occurring or anthropogenic agents. There are four main components of a risk assessment. Th ese include (1) risk identification, (2) doseresponse assessment, (3) exposure assessment, and (4) risk characte rization (Paustenbach 2000). Risk identification is the first sub-process of risk assessment whereby a determination is made as to whether exposure to an agent can cause an increase in the incidence of a particular adverse health effect (e.g., cancer, birth defect) and whether the adverse health effect is likely to occur in humans (Pau stenbach, 2000). Surveys and sampling are the two main methods employed in risk identification.
20 Dose-response evaluation is the process of quantitatively eval uating the toxicity information and characterizing the relations hip between the dose of the contaminant administered or received and the incidence of adverse health effects in the exposed population. From this quantitative dose-res ponse relationship, t oxicity values (e.g., reference doses and slope factors) are derived that can be used to estimate the incidence or potential for adverse effect s as a function of human exposure to the agent. These toxicity values are used in the risk characterization step to estimate the likelihood of adverse effects occurring. Exposure assessment is the third step in a risk assessment and is a qualitative or quantitative description or estimate of th e magnitude, frequency, duration, and route of exposures to the various populat ions (Paustenbach, 2000). This process considers several factors before estimating the absorbed dose, such as exposure duration, exposure route, chemical bioavailability from the contaminat ed media and the specific physiology of the population. Knowledge of the ch emical concentration is n ecessary for the accurate estimation of the absorbed dose. Risk char acterization, the final component of risk assessment describes the nature and magnitude of risk, including uncertainty (U.S.EPA, 1996). Risk characterization includes both quantitative estimates an d qualitative descriptors of risk, as well as discussions about key model assumptions and data uncertainties (Williams et al, 2002). Unlike the other steps of the risk assessment process, risk characterization cannot and does not solely rely exclusivel y on guidance documents and formulas to capture the importance of an alysis. Largemouth bass from the Hillsborough River, Tampa Bay contained 0.56 ppm mean wet weight of total mercury (Howard et al in press 2008). When an exposure rate of 365 days for a lifetime of 70 years, at a given ingestion rate of 30 g/day is considered, these levels exceed the Environmental Protection Agencys safe oral reference dose of 1E-4kg/day (USEPA IRIS, 2008). The reference dose is determined by using the following equation derived by the USEPA:
; M FUF NOAEL RfD where RfD Reference Dose is the acceptable level of risk that should not cause any adverse effects. The chronic oral reference dose for methyl mercury was a 1E-4/kg/day (USEPA IRIS, 2008). NOAEL No Observable Adverse Effect Level (N OAEL) would be the driving force in determining the level that should be allowe d as it does not cause an adverse effect in those individuals. A safety factor, using an uncertainty factor (UF) and a modifying f actor (MF) are applied to the chronic NOAEL for methyl mercury. UF is a factor of safety applied to the reference dose and takes into account th e difference between people and the use of laboratory animals. MF is a margin of safety used to take into account if there is a lack of professional knowledge and judgment about the chemical. An uncertainty factor of 10 was used and a modifying factor of 1 was applied in the calculation of the chronic reference dose for methyl mercury (USEPA IRIS, 2008). Contaminated fish is the main source of methyl mercury, approximately 95% of methyl mercury in fish is readily absorbed in the gastrointes tinal tract (Counter and Buchanan, 2004). Pregnant women and children have been identified as higher risk groups. Maternal daily dietary intake levels were used as the dose surrogate for the observed developmental effects in the children exposed in utero. The reference con centration or dose measure methyl mercury is based on the assumption that thresholds exis t for certain toxic effects such as cellular necrosis (USEPA, 2008). Mercury is noted by many researchers to be a neurotoxin and is linked to sensory, immune, motor, and neurological dysfunc tion (Bernard, et al., 2001; Counter and Buchanan, 2004; McDowell, 2004). Minamata disease was first discovered in the Yatsushiro Sea coastal area, particularly around Kumamoto Prefect ure, Japan in 1956 where it was used as a catalyst in acetaldehyde production. This became the first well documented case ( Minamata Disease Research Group; 1968 Harada M; 1994 ) relating industrial use of mercury and its biomagnifi cation through the food chain (Herada, 1994). 21
22 Total mercury and methyl mercury bioaccumulated in the local fish stocks and residents were accustomed to eating 17.6 ounces of fish per day from the Shiranui Sea. A study conducted 50 years later in Mimimata, reveal ed that persons were still experiencing neurological symptoms and sensory disturbance although the me rcury levels in the hair were normal (<10ppm) (Herada et al., 1998). This may be due to the long elapse of time between the peak mercury exposures and samp ling time, and hair may not have been a useful indicator. Many studies have associated mercury toxicity with other routes of exposure aside from the consumption of fish, such as occupati onal exposures, vaccinations, and religious practices. For example, in the 1800s, feldspar hat manufacture work ers were exposed to hazardous levels of mercury vapor and experi enced symptoms such as trembling, also known as hatters shakes, slurred speech, loss of coordination, memory loss, irritability and anxiety (Connealy, 2006). This would be later called Mad Hatters Disease. Bernard, et al (2002) noted that 1 in 150 children in Un ited States have autism and linked the epidemiological incidence of the regressive form of autism to the use of thimerosal, a preservative containing mercury used to vaccinate young children below the age of two. Riley, et al (2006) determined that mercury was be ing used for religious practices among the Latino and Caribbean populations in United States. However, fish consumption still remains the major exposure route for merc ury toxicity, especially among coastal populations such as Tampa Bay, Florida. Acco rding to Mahaffey et al. (2008), women in coastal regions (resident within 25-50 miles of the coast) had higher blood mercury levels ( g/kg) relative to their ne ighboring inland regions. Consuming more than the recommended adviso ry guidelines does not necessitate that all persons experience the same health implicati ons. A number of factors determine whether a given person experiences adve rse health effects associated with the indiscriminate consumption of mercury in fish. These fact ors include but are not limited to; age, sex, genetics, individual excretion rates, individual absorption rates in the gastrointestinal tract, concentration of merc ury consumed, amount of fish consumed, frequency and
23 exposure time. Fish advisories are establis hed to protect public health. This study determined if subsistence and recreational fish ermen had seen local fish advisories and if they had seen this information, did the info rmation facilitate a change in individual behavior. This study utilized selfreported survey data to reveal risk perception of mercury contaminated fish. 3.3 Perceived risk Risk perception refers to the body of rese arch tapping into peoples recognition and concerns about risks (Slovic, 2000). Many studies have shown how different factors have influenced risk perception in cluding gender, income, leve l of education, family and friends perception, experience and age. A semi-structured survey geared towards understanding the interac tion of these variables and their contribution to the awareness of mercury risk associated with fish consum ption among recreational fisherfolk in the Tampa Bay region would be very useful in de termining which of the outlined factors are more influential in this study. There may be large differences between cal culated risk determined by toxicology and statistical analyses and the perceived risk by laypersons. Toxicologists place great emphasis on exposure and dose-response when evaluating chemical risks; however laypeople tend to believe that any chemical exposure is detrimental to human health (Slovic, 2000). Agencies may argue that the reference dose is devel oped to protect even the most sensitive persons in the population and that once persons follow guidelines, there should not be any health implications While this may be true, average persons would not know what a reference dose is and how they are expected to follow the guidelines. Humans have always been intuitive toxicolo gists, relying on their keen sense of sight, taste, and smell to detect unsafe food, water, and air (Slovic, 2000). However, mercury does not have a detectable smell or taste, and its most toxic form, methyl mercury
24 (MeHg) is associated with the protein and fa t in fish. The long term dangers presented with the consumption of contaminated fish is eminent to persons in the risk profession, but to recreational and subsistence fishermen, who derive nutritional benefits from the consumption of local fish, may attenuate indivi dual risks and vulnerabi lity factors. A fish advisory is only useful if average persons are knowledgeable of the risk and understand the risk associated with the indiscriminate consumption of contaminated fish. Department of Health and Human Services and USEPA argue that there are higher risk subgroups within a given population, such as women of childbearing age, pregnant and nursing women and young children which are more vulnerable than ot her demographic sectors. Several studies have now published on the pot ential adverse effects on child development associated with prenatal exposure to me thyl mercury (Clark son and Magos, 2006). Recent literature suggest s consumption of more than 1.23 oun ces of fish per day of large predatory fish may actually increase mens ri sk to coronary heart disease (Etherson, et al ., 2003). However, state and local agencies do not recognize men as a sensitive group, to be highlighted in consumption guides. Krosknich, (1999) noted that most ethnographi c studies sampling e fforts are focused on obtaining results from those with lower le vels of education. However, young and old males and persons with the highest income levels are underrepresented in ethnographic studies. It was argued that ge nerally white males may perceive less risk than others because they are more involved in creati ng, managing, controlling and benefiting from technology, as well as hazardous activities (Slovic, 2000). Slovic,(2000) stated: Women and non-white men may perceive greater risk because they tend to have less control over these activ ities and benefit less from them (pp.399-400). A national study estimated that high income i ndividuals accounted for more than 25% of the national licenses purchased. It is anticipated that the more affluent licensees and those working long rigid hours would be more amen able to online surveys. The combination of
25 convenience field and random online sampling pr oduced greater insight into the advisory awareness, risk awareness and percepti ons of mercury contaminated fish. Without a doubt, risk perception seems to be re lated to an individuals power to influence decisions about the use of hazards (Gustafson, 1998). In this study, a persons power to influence their individual ri sks could be influenced by their occupation, income and political association. In this study, occupational vulnerability and income are factors included in the advisory awarene ss (information viewing) model. Kasperson (1988) notes that ther e are four factors of risk influencing risk perception: voluntariness, ability to influence the circum stances surrounding the risk, familiarity with the risk and catastrophic poten tial. Figure 3.1 shows the distri bution of perceived risk of 81 hazards using psychometric analysis of questionnaire data derived from Slovics, (1987) study. Mercury can be found in top righ t corner, as its risk s are involuntary and mercury can not be visibly seen and ill effects not well known. Slovic, 1987 believes voluntariness of risk exposure and trade-offs between hazards and benefits (acceptability of risks) play a crucial role in risk perception. It was not ed that people are willing to accept risks 1000 times as great from a volunt ary activity as they would from an involuntary activity.
Figure 3.1: Risk perception of 81 hazards investigated (Slovic, 1987), pp. 5 Public perceptions are a product of innate biases and economic interests by different entities. It was also noted that risk percepti on does not usually mimic the true nature of the risks. According to Kasperson et al ., (1988), amplification occurs at either the transfer of information about the risks or in the response mechanism of society. Signals concerning risks are refined by individuals and entities who communicate these risks. These entities include agencies, the news me dia, cultural groups, in terpersonal networks 26
27 and scientists. Behavioral responses are considered a secondary effect of risk amplification. In this study, th e role that agencies and me dia play in recreational and subsistence fishers heeding the advisories was examined. Freudenberg (1992) examined factors contri buting to the attenuati on of risks signals within agencies. These factors include: la ck of organizational commitment to risk management function, specialized divisions of labor that create corporate gaps and the bureaucratic attenuation of in formation flow within the organization. The media plays an important role in establishing if people take notice of the risk s and heed the advisories or warnings that are communicated. Risk atte nuation has the potential of being more dangerous as persons are quite comfortabl e, although the situation warrants some apprehension. Amplification of risks could be financially damaging in many cases. This is especially true for those in the food indus try. Stigma of risks is used by risk researchers to refer to the generalized ne gative image of a technology, place or product being overly dangerous. Stigma tized technological hazards are often associated with dreaded consequences and involuntary exposur es. The notion of stigma violates the natural order or any just sta ndards. Impacts connected with the risks tend to be unevenly distributed.Many studies determined the s ources of public health information and deduced which groups were at greater ri sks due to the lack of awareness and misconceptions of environmental hazards (Burger, 2000a; Sustein, 2002). In a risk perception study on mercury fish advisories which examined sources of information, perception and compliance, Burger et al (1999), established that: 64% of the sample population obtai ned fish and fishing information from other fishermen or from bait and tackle shops (38%), rather than magazines, radio, television, Department of Environmental Protection newsletter or brochures, the New Je rsey Health Department or their own doctors (pp. 221).
28 A riskcommunication study of radon revealed that people hold many accurate beliefs about radon exposure, in addition to many inaccu rate beliefs. Many believed that radon contamination is permanent. Few people (13%) understood that radon decays quickly (Sunstein, 2002). In this cas e, the participants beliefs and the lack of knowledge pertaining to radon exposure made the problem seem more severe and unsolvable than the reality of the risk. This study determined that there were si milar common misconcep tions concerning the fate of mercury in the envir onment and health risks. A pr eliminary study revealed that there were no posted advisories at the most visited publicly accessed fishing areas and recreational parks along the Hillsborough River in the Tampa Bay region. It is uncertain as to how many persons understand the risk of mercury in fish and if supplemental guides are necessary to target unaware fishers and consumers. 3.4 Sustainability as it relates to risk perception of mercury in fish Prior to the 1970s, there was a common belief th at most resources-water, soils, forests, including fish were inexhaustible (Sigler, 1984; Freese, 1997). Th is notion has changed as there is a greater understand ing of fisheries as both an ar t and science. Scientists and most of humanity recognize th at there is a limit or carryi ng capacity to any population of a species or natural resource. Exploitation of this resource is constrained by its natural growth and recruitment but coordinated utilization of the resource by the va rious agents can ensure its sustainability (Bischi and Lama ntia, 2007). Agents include all users of the resource and stakeholders, such as recreational and commercial fishers, environmental groups and agencies. Good stewardship a nd wise environmental management are necessary to ensure a balance is maintain ed in an ever demanding world, where the integrity of the resource is at stake (Sigler, 1984). There are many views as to the best appr oach that should be used when managing fisheries and increasing sustainability. The mo st common and practical approach used is
29 the positivist approach, which uses scientific data in the advancement of models and fishery management theories. A harvesting time-discrete model compares constant fishing effort and profit maximization methods and includes external ity and equilibrium concepts; results showed that with oligopol istic competition, even if a preserve was adjacent, persons acted hysterically and s howed irreversible unsustainable behavior (Bischi and Lamantia 2007). In this case th e precautionary principle towards fisheries should be adopted as stocks would be diminished below levels of any possible restoration. The precautionary principle is a notion devel oped in response to risk uncertainty and emphasizes health or environmental maintena nce (Precautionary Principle Project, IUCN, 2003). Sustainability is related to the precauti onary principle, which states when there are threats of serious or irre versible damage, lack of scie ntific certainty should not be used as a reason for postponing measures to prevent environmental degradation. (McPhee, 2008, pg 59). This approach has generated mixed sentiments among developers, researchers, stakeholders, a nd the local and academic community. Many argue that this approach may lead to co stly reduction of harmless pollutants and a tolerance for strong negative effects, if re duction methods are not available (Jaeger, C.C et al, 2001). Whilst others disput e that development brings risk s as well as benefits, it is not unreasonable to demand solid evidence of safety. In reality, the precautionary principle may shift the burden of proof from those who want to avoid risks to those who want to take risks, much as the burden of pr oof lies with the prosecution in criminal trials or with the hypothesis in statistical tests (J aeger, C.C et al, 2001). Many researchers (Kannan and Falandysz, 1998; Karouna-Renier et al ., 2008) have been able to quantitatively demonstrate that Fl orida faces a long-ter m peril from mercury contamination. In addition, the Florida Depart ment of Health (FDOH) has examined the trends of anthropogenic mercury mass flow and emissions in Florida and results revealed that coal combustion contributed 136,400 pounds to the total cumulative mercury mobilization and contributed to 50% of the emissions. However, few studies have
30 integrated environmental, socio-economic and cultural factors in analyzing fishing behavior, consumption and risk perception of mercury contaminated fish in the local setting of Tampa Bay. The expected outcomes of the study are: a) increased local awareness of consumption risks connected wi th local species of special concern, b) a better understanding of the need to have more access to mercury related information, c) more effective methods of communicating ri sks associated with mercury and fish consumption, and lastly d) an understanding of the delicate balance between consumption risk and fishery resources management. Increasing risk knowledge and communication of mercury associated risks with the consumption of fish may have implications fo r sustainability of the resource. Most game fish in the Tampa Bay region, whether freshw ater or marine are usually the large, predatory fish such as the largemouth bass, King Mackerel and Shark at the top of the food chain. These fish have high levels of mercury associated with them (Florida Department of Health, 2007). With increased effective communication, persons will have a greater understanding of the risk associated with the consumption of larger, predatory fish and may begin catching smaller fish or fish at lower trophic levels, such as perch in an effort to reduce mercury related risks. Th is may be encouraging from a public health standpoint; however this may cause an ecologi cal imbalance if the natural dynamics and carrying capacity of individual species are surpassed. If the predatory fish at the top of the f ood chain are being caught for game, there are usually no higher consumers other than man and carnivorous birds. However, if a scenario exists in the future, where fishers be gin targeting smaller si zed fish of the same game species in an effort to reduce hea lth risk, the fish populations may go below a critical level or carrying capaci ty. Carrying capacity refers to the number of individuals or population that can be supported in a given area within natural resource limits, without degrading the natural, social cultural and economic environm ent for present and future generations (Urban Environmental Ma nagement, 2008). Currently, freshwater recreational permit requirements have a maximu m size restriction of 14 inches, but there
31 is no minimum size restriction for bass in sout h Florida, so fishers may capture immature fish before given the opportunity to spawn, in an effort to reduce individual health risks. In addition, if consumption rates of fish at lower trophic levels are beyond the natural capacity of the given species population, the higher trophic level consumers would be affected and the natural ecological dynamics may be offset. A situation may arise where there are not enough primary and secondary level consumers to support the larger predators (Primack, 2004) (Meffe et al, 1997). Many studies have examined factors influencing the carrying capaci ty. This definition suggests that a sustainable approach needs to be adopted, if a population is to survive. Overshooting refers to an area in which the fish stocks are being reduced and not being replaced at the same rate, t hus the resource is not being ut ilized sustainably (Figure 2.9). Fishery management considers many variables in determining the he alth of a population stock or carrying capacity: abundance (numbers or weight), recruitment, food habits, movement, age composition and growth rate, structure and mortality rates (disease, parasites and stress), yields, size, reproductive rates and relativ e catch/ effort (Sigler et al, 1984). As previously mentioned many studies utilize many factors in their model to determine the carrying capacity. Two study ar eas may have different carrying capacities due to varying reproduc tive rates (Bischi et al 2007). Likewise management practices, such as the presence of preservation which prohibits catch may increase recruitment, reduce disease and encourage overall ecological health (Trisak, 2005). Population Carrying capacity Exploitation Overshoot Where, Exploitation >carrying capacity Time (years) Figure 3.2: Graphical representation of carrying capacity and overshoot concept.
32 Chapter 4: Research Design 4.1 Introduction This chapter will provide insight as to the re search objectives, the research questions that are being answered and the approach that was adopted in answering these questions. The methodology could be divided into three main sections: development of a survey, data collection and data analysis. 4.2 Research Purpose The objectives of the study include: To understand which factors (ethnicity, le vel of education, age, sex, sources of information) influenced vi ewing of fish advisories To determine if there is adequate di ssemination of mercury related risk information by local and state agencies to recreational and subsistence fishermen in the Tampa Bay Region. To contrast mercury risk knowledge be tween field and online participants To increase community awareness of merc ury risks associated with locally caught fish by providing participants the E nvironmental Protection Commissions consumption guide. It is anticipated that this information will enable those who had not seen local fish advisories to make informed decisions in the future.
33 4.3 Research Questions Research Question 1: Are mercury consumption risks associated with fish being communicated effectively by environmen tal policy makers (state and local government)? Research Question 2: Do all fisherfolk have equal access to, and understanding of risk consumption information that al lows for informed health decisions? Research Question 3: Does mercury ri sk information change behavior among fisherfolk? The first research question was answered by ex amining the data gathered from online and field surveys and informal interviews. In addi tion, a review of the legislative literature and in-depth key agency pers onnel interviews were conducte d to determine the efforts made by agency officials to educate the loca l fishing community of mercury consumption risks. The second question determines whic h independent variables influenced risk information viewing, a prerequisite for me rcury understanding and awareness. This question will be answered by examining source s of public health info rmation utilized and an analysis of open-ended risk understanding questions. In addition open-ended responses from field and online surveys will be used to examine access to local fish advisory information. The final research que stion examined reported behavior change and if evidence suggested a change, how the behavior had changed and determining if there was any known scientific merit in the individuals change, such as capture and consumption of smaller sized fish and change in desirability of certain species of fish. This data will be derived by field and online surveys.
34 Chapter 5: Study Area 5.1 Introduction For determining access to local fish advisory information and the risk perception of mercury contaminated fish among the recreational and subsistence fishermen, a case study approach was adopted. Five study sites were utilized by the principal investigator in the field portion of the study. However, onl ine surveys recruited licensees from the coastal counties of Hillsbor ough, Pinellas and Manatee within the Tampa Bay region. 5.2 Site selection The five study sites included: Rowlett, Lo wry Park, Ballast Point, Gandy Bridge and Tarpon Springs offshore excursion (figure 5.1). Th ese five sites were chosen due to their popularity among local subsiste nce and recreational fisher men and the areas being designated by USEPA as being contaminated due to mercury. This study focused on risk perception of mercury contaminated fish among local subsistence and recreational fishermen in Tampa Bay; hence it was necessary to find fishing locations that were easily accessible to the public. Florida residents fish ing off public jetties and docks are exempt from Florida Fish and Wildlife Conserva tion Commission requirements of needing to possess a license. One of the questions on the survey asked participants if they were a resident of Tampa Bay (Hillsborough, Manatee and Pinellas Counties). Local fishermen were targeted in this study, as fishing beha vior and consumption patterns were examined in those who had access to fishery resources all year. It is expected that these persons probably have greater health risks associated with cons umption of locally mercury contaminated fish.
Figure 5.1: Study sites showi ng field convenience sampli ng and online random survey resident area 5.3 Physical context Tampa Bay is considered a natural harbor al ong the Gulf of Mexico on the western coast of Florida and comprises of Old Tamp a Bay, Hillsborough Bay, McKay Bay, and New Tampa Bay. It is bounded by Pinellas County on the west, Manatee County on the south, and Hillsborough County to the east of Pinellas County. Hillsborough County is the largest of the Tampa Bay Metropol itan Statistical Area (Table 5.1). 35
Table 5.1: Area of Counties in the defined study area (Florida Geographic Data Library data) County Area (Sq. miles) Hillsborough 1070.869 Manatee 763.107 Pinellas 289.178 This region contains Florida's largest ope n-water estuary (>398.071 square miles). In addition, these three counties share access to the Bay and Gulf of Mexico. The Bay is popular for sports and recreation and supports one of the world's mo st productive natural systems (Tampa Bay Estuary Program, 2008). It is estimated that approximately 64% of recreational boaters surveyed in Tampa Bay between 2003 and 2004 were engaged in fishing (Sidman et al ., 2004). It has been estimated that values of access to the regions fisheries is $39.52 per angler per year which suggest that a net value of $8, 335, 710 per year (Greene et al, 1997). The popular fish ing areas within Hillsborough County include Lake Thonotosassa, Edward Medard Reservoir and Hillsborough River. Some of the most popular freshwater sites in Pinellas include Lake Seminole and Tarpon Springs. Some of the popular fishing sites within Manatee County include Lake Manatee and Manatee River. More than 200 species of fish are found in the Tampa Bay region. The popular freshwater fish in the Hillsborough River include largemouth bass ( Micropterus salmoides), Redear Sunfish (Lepomis microlophus), bowfin ( Amia calva ), and gar ( Lepisosteus osseus ) (Karouna-Renier et al 2008). A considerable portion of the study area has been deemed impa ired by the Florida Department of Environmental Protection whic h suggests that there is reason for some consumption safety concern (Table 5.2). Cl ass 3 used for recreation, propagation, and maintenance of a healthy, wellbalanced popul ation of fish and wildlife had the greatest areas of impairment in all three counties (figure 5.2). Class 2 used for shellfish propagation or harvesting was the second la rgest impacted class. Hillsborough and Manatee Counties possessed some class 1 impair ed, which accounted for less than 1% in 36
the former and approximately 6% in the latter. Figure 5.2 shows all the areas within Hillsborough County that are considered impa ired due to mercury contamination. Figure 5.2: Areas being classified by Florida Departme nt of Environmental Protection (FDEP) as being impaired 37
38 Table 5.2: Proportion of area deemed as impaired County Area considered impaired Sq. miles Area not considered impaired Sq. miles % of County Area classified as being impaired Hillsborough 129.450 941.419 12.09 Manatee 190.934 572.172 43.09 Pinellas 129.487 159.691 44.77 The popularity of freshwater and saltwater fi shing within the regi on coupled with known mercury contamination of this geographic area s sediment and fish pr ovides context as to why determining fishers awareness of mercury contamination in local fish is so crucial. This study will examine awar eness of mercury among licensed fishermen/fisherwomen within the study area. Semi-structured fiel d surveys will be conducted to ascertain a greater understanding of individual perceptions related to mercury consumption risks. 5.4 Social context According to the Hillsborough County Pla nning Commission, (2007) the population of Hillsborough County is approximately 1.2 millio n and is the most populous in the Tampa Bay Tampa-St. Petersburg-Clearwater Metropol itan Statistical Area, followed by Pinellas County. Tampa Bay, the second largest metropolitan area in the State of Florida, has a population of approximately 2,697,731 (State of Florida, 2006) with an average annual population growth of approximately 2.47%. There are three prominent cites along the bay: Tampa, St. Petersburg and Clearwater. Educational attainment se emed to be similar for all three counties, with the largest proportion being hi gh school graduates (>30.0%). Although ethnic composition was quite diverse, the majority (81.3%) were Caucasian (United States Census Bureau, 2000). Fl orida median annual income is $41,226. The median annual income was used to determine how the income categories were characterized. Three categories were utilized: below media n, median and above median levels of income.
39 Table 5.3: Educational atta inment of the population censu s for the respective counties Educational attainment Hillsborough Manatee Pinellas Less than high school graduate 18.0% 17.41% 13.2% High School graduate 30.0% 34.76% 31.3% Some College or associates degree 27.0% 20.64% 30.5% Bachelors degree or higher 26.0% 27.19% 25.0% 5.5 Summary The field surveys were conducted at a total of five highly visited fishing sites among local fishermen that were id entified by USEPA as being impa ired. In addition, resident licensees in Hillsborough, Pinell as and Manatee Counties were recruited to participate in the online survey. The next section outlines the methodology employed in this study.
Chapter 6: Methodology 6.1 Introduction The methodology comprised of the survey proc esses and the semi-str uctured interviews with key personnel from agencies. The former is outlined in section 6.2. Semi-structured interview methodology is outlined in section 6.6. 6.2 Introduction to surveying sampling method The survey method is divided into 3 main sec tions: 1) development of a survey, 2) data collection and 3) analysis The latter is described thoroughly below. Figure 6.1: Survey design and analysis schematic for survey methodology. Summary of survey methodology Development of a Survey Analysis The response data was coded. Descriptive statistics and participant responses were used to summarize results. Binary lo g istic re g ression determined access to fish advisories amon g Data Collection Convenience sampling at 5 locally accessed fishing areas (field survey). Entire sample of fishing license holders (online survey). 3 categories of questions: socio-demographic, fishing consum p tion behavior and fish health safet y 40
41 6.3 Development of the Survey The development of the survey included so cio-economic, fishing consumption behavior and fish health safety questions. The survey comprised of 22 questions: structured and open ended. Socio-economic questions determ ined a respondents sex, age, household income and occupation. Fish consumption beha vior questions were posed to determine overall fish consumption patterns and if partic ipants consumed local fish, what fish they consumed and how often. Human knowledge of overall fish health safety was gauged by determining if subjects had (1) known me rcury was found in some fish and was dangerous to humans; (2) previously seen information mentioning that mercury was found in certain fish more than others and; (3 ) the impact of the fish advisory information on consumption behavior. 6.4 Data Collection Two surveying sampling methods were utilized in this study: field convenience sampling and online invitation recruitment using Florida Fish and Wildlife Conservation Commission (FFWCC) license database w ith Hillsborough, Manatee and Pinellas residents. Upon completion of the survey, an Environmental Protection Commission (EPC), Hillsborough County fish advisory (Figur e 4.1) was given to field participants to increase awareness of mercury in fish. Th e fish advisory highlights three groups at greater risk: children, pregnant women and wo men of childbearing ag e and indicates that their consumption should be less than that of the typical adult. Likewise, online respondents were given a web link to the Flor ida Department of Health state advisory with species of special concern for the various geographic locations (figure 4.2).
42 6.4.1 Online data collection The Florida Fish and Wildlife Conservati on Commissions database was acquired in October 2008. The zip codes were sorted a nd zip codes corresponding with Hillsborough, Manatee and Pinellas were selected. Th ese addresses were then imported into ArcMap9.3. These were then geocoded to veri fy the residences fell within the specified counties. The successfully geocoded addresses we re then exported and used to create an email address book in surveymonkey.com. Di fferent address books were created based on the county and type of license. A total of 28 license databases were created with a combination of Pinellas, Hillsborough and Ma natee Residents. A complete listing of these databases can be found in appendix 2. Three selection criteria were us ed for online surveys: 1) th e participant was successfully geocoded, 2) the participant had to have an ema il contact listed within the database and 3) the participant had to be older than 18 years of age at the survey date. An invitation email was sent from surveymonkey.com. The invita tion was entitled Fis hing in Tampa Bay explaining that the survey was examining the use of fishing as a resource, the expected time duration of the survey and a link wa s provided to commence or opt out of the survey. Appendix 3 shows the em ail sent to the recruitee. 6.4.2 Data utilized in binary logistic regression Florida Fish and Wildlife Conservation Co mmission licenses successfully geocoded with email addresses were used to create an email list serve for licensed fishermen recruitment. The survey was distributed between November 2008 and September 2009. Most of the responses from the socio-dem ographic, fishing behavior and health information sections were categorical and were coded. The names and definition used in this study are summarized in Table 6.1.
43 Table 6.1: Summary table with definitions and categories used in binary logistic regression analysis of online survey data Variable Definitions and categories us ed for binary logistic regression Ethnicity White=0, Not White=1 Sex Female=1, Male= 0 Household income Below median(less than 40,000 per annum), Median(40-60,000), Above median (more than 60,000) Fishing frequency Participant fishes more than once per month=1, Participant fishes less than once per month= 0 Occupation Vulnerability Occupation is within an industry iden tified by FDOH as having a higher mercury vulnerability=1, Occupation is not within an industry identified by FDOH as having a higher mercury vulnerability=0 Highest education level attained Categories: Vocational, Some High School/ high school, Some college/ college graduate, Post-graduate Types of licenses Categories: freshwater only, saltwater only, both saltwater and freshwater Fish Consumption Participant and/or family eats fish caught =1, Participant and/ or family does not eat fish caught=0 Sources of public health information Internet sources=1, Non-internet sources=0 Participant reported seeing information mentioning that mercury is found in different types of fish Yes, seen information=1, No, had not seen information=0 6.4.3 In-field Data Collection Convenience sampling was used to conduct a survey at five popular publicly accessed fishing areas; Ballast Point, Gandy Bridge, Rowlett, Rotary Park and Tarpon Springs between November 2008 and August 2009. A reconnaissance of the fishing areas revealed that anglers had a preferred fishing time of wee kday evenings and weekends. Most surveys were conducted at these times. Two selection criteria were used when conducting the field survey: 1) participants had to be actively fishing on the date in which they were approached and 2) survey respondent s had to be older than eighteen years of age. It should be noted that most of th e fish caught and eaten were listed on the consumption guide (figure 2.4). These consum ption guides were given to participants upon completion of the survey. Attention was draw n to the fish they identified as being consumed and the level of risk associated with the species and th e geographic location.
44 6.5 Analyses Although the written survey instrument was th e same, the data obtained from online and field surveys were analyzed separately, as these utilized different sampling methods. Online surveys utilized an indiscriminate random approach to model the licensees; however the field survey facilitated explanat ory results. A comparison of the participants consumption risk knowledge and behavior fr om online and field methods was conducted to determine if these could be deemed as two different socio-demographic segments of the Tampa Bay fishing populations. 6.5.1 Online survey analysis The structured responses were also coded using the method described in section 5.2. A binary logistic regres sion was used to determine if nine socio-demographic and fishing behavior independent variab les influenced people viewing mercury related risk information associated with the consumption of locally contaminated fish (dependent variable). It should be noted that the view ing of mercury advisories was selfreported and none of the information disclosed was independently verified. Two z-test for proportions test were conducted. The first z-test for proportions was conducted to determine if reproductive women viewing of fi sh advisory information was significantly different to women in all other age cohorts. The second z-test for proportions was used to compare percent of female respondents to the regional number of female recreational licenses sold. Variation Inflation Factor (V IF) determined if mulitcollinearity between independent factors was of major concern. 6.5.2 In-field survey analysis The responses from the structured demogra phic portions of the survey were coded to make subsequent analysis easier and more efficient. Descriptive statistics was used to summarize the data obtained. Field notes were analyzed to determine if there are trends
45 among the parks and if there are differences between the responses or behavior observed. Responses from the open ended questions were then analyzed to determine if persons truly understood the risks and were using an e ffective means of reduc ing individual risks. In many cases verbatim quotes will be used to express key findings. 6.6 Key informant semi-structured interviews Seven main questions or topics relating to mercury issues were prepared to ask agency officials (appendix 5). It should be noted th at questions were not limited to these seven questions and responses from the prelimiary questions facilitated more discussion. This technique was utilized to gain an in-depth insight into motives of restrictions and laws governing mercury policy and fish advisories.
46 Chapter 7: Results 7.1 Introduction The results have been divided into three main sections: 1) online survey results, 2) field surveys, direct and participant observations results and 3) qualitativ e results derived from interviews with agency officials. 7.2 Online survey results The overall response rate for the online su rveys was 5.58% when c onsidering the total population (N=7334) of licensed fishermen with email contacts. This population comprised of a combination of databases created for Hillsborough, Ma natee and Pinellas counties. It should be noted that individual database res ponse rates varied widely. For example, some databases with lifetime licensees and military sportsmen had a zero percent response rate, whereas the response rate for a database with chartered licensees had a 100% response rate. It should be noted that 13 out of office notifications were received by the principal inve stigator. Only completed onlin e responses were considered in the descriptive statistics and binary logist ic regression portions of the analyses. Table 7.1 shows the descriptive statistics for the variables considered.
47 Table 7.1: Descriptive statistic s for variables analyzed Self-reported viewing of me rcury related consumption risk information (N= 375) Variable: Categories: (if applicable) Mean SD Age Young adult (18-50) 0.632 0.483 Middle age adult (51-65) 0.333 0.472 Seniors (>65) 0.032 0.176 Sex Females=1, Male=0 0.123 0.328 Ethnicity Non-whites=1, Whites=0 0.101 0.302 Education Vocational 0.051 0.220 Some high school/ high school graduate 0.165 0.372 Some college/ College graduate 0.587 0.493 Post-graduate tertiary education 0.189 0.392 Income Below median 0.096 0.295 Median 0.159 0.367 Above median 0.743 0.437 Fish consumption Yes/ sometimes=1, No= 0 0.880 0.325 Fishing frequency >1/ month= 1, < 1/ month=0 0.627 0.484 Internet sources Internet Sources=1, Non Internet source=0 0.669 0.471 Possible occupational vulnerability Possible vulnerability= 1, not probable=0 0.216 0.412 License type Saltwater only 0.221 0.416 Freshwater only 0.456 0.499 Both freshwater and saltwater 0.323 0.468 Vi ewing of information Viewed information=1, Did not view information=0 0.853 0.354 Results from the online survey showed that most of the participants were young adults (18 and 50), male (87.7%) and Caucasian (89.9%). The majority (77.6%) of the participants had received tertiary college education. Most (74.3%) had an income above the Floridian median ($40,000 to $60,000 per annum). Most of the participants (85.3%) stated that they viewed health informa tion from the internet. Not many could be considered to have a job that fell within an industry classified as having an occupational vulnerability (21.6%). Approximately 63% of the respondents fish ed at least once a month and could be considered regular fishermen or fisher women. The largest pr oportion (45.6%) of respondents had freshwater licen ses. 88% of online respondent s indicated that they or their family ate the fish they caught. 85.3% of online participants stated that they had seen information mentioning that mercury was found in certain types of fish. The
48 independent variables listed above were used in a binary logistic re gression to determine which factors influenced viewi ng of mercury risk information. Table 7.2: Logit coefficients and odds ratios from multivariate analys is of self reported viewing of mercury related consumption risk information Self-reported viewing of me rcury related consumption risk information (N= 375) Variable Categories (where applicable) Co-efficient Std. error Odds ratio Age Young adult -0.681 1.075 0.506 Middle aged adult -0.443 1.096 0.654 Sex -0.941** 0.400 0.390 Ethnicity 0.123 0.579 1.131 Education Some high school/ high school -0.124 0.713 0.884 College -0.168 0.654 0.845 Post-graduate degree 0.460 0.761 1.584 Income Median 0.071 0.522 1.073 Above median 0.688** 0.448 1.990 Internet sources 0.558* 0.318 1.748 Possible occupational vulnerability -0.458 0.354 0.632 Type of license Freshwater only -1.116** 0.519 0.328 Both Freshwater and Saltwater -1.145** 0.540 0.318 *p < 0.10. ** p < 0.05 In this study, viewing of mercury consum ption information is a pre-requisite for understanding risk. One must have seen mercury risk information or a fish advisory before one can understand consumption risk. Mu ltivariate logistic re gression was utilized to investigate the statistical effects of age, sex, income, the use of internet sources for public health information, possible occupa tional vulnerability and license type on viewing of mercury related consumption risk information. The results of this regression are shown in table 7.2. The Nagelkerke Rs quare value was 0.114. Results revealed that education, age, ethnicity and occupation vari ables did not significantly influence the viewing of mercury related information. Table 7.3: Comparison of socio-demographics of participants and the license purchases Variable tested Scale Online Purchased Z score N % N % Women Region 45 12.3 592456 13.5 1.042 Above levels of income National 279 74.3 7049767 24.8 60.6*** *** p<0.01
49 Sex of the participant did influence the vi ewing of mercury risk information. Being female influenced the viewing of mercury risk information (p< 0.05), when males were used as the reference variable. There were more male than female respondents (12.3%), however the Zscore of this studys proportion of female respondents was not significantly different (p< 0.01) when comp aring proportions of regional licenses purchases (American Sportsfishing Associat ion, 2005) (Table 7.3). In addition, results also indicated viewing of risk information by women of reproductive age (18-50), a target group for fish advisories was not signif icantly higher (zvalue=1.042, p>0.10) than women of all other age cohorts. The use of internet sources to obtain public health information was significant at the 0.10 level. Having an above median level of in come was considered significant (p< 0.05) in influencing the viewing of mercury relate d risk information, when having a below median income was used as the reference comp arison variable. It should be noted that the proportion of participants with an above medi an level of income was significantly higher (z-score=60.6, p<0.01) than the national proporti on of persons within this income bracket (table 7.3). However, having a median in come showed no significant difference in influencing viewing of mercury related information, when possessing below median income was used as the reference comparison variable. Results from the binary logistic regression also suggested that the type of license also influenced if the mercury consumption risk information was viewed. Having a freshwater only license was considered significant (p< 0.05) in influencing the viewi ng of mercury related risk information, when possessing a saltwater licen se was used as the reference comparison variable. Likewise, possessing both freshwate r and saltwater license was considered significant (p< 0.05) in influencing the viewing of mercury related risk information, when possessing a saltwater license was used as the reference comparison variable.
50 Table 7.4: Summary table contrasting t ype of license and level of income Level of income Type of license Saltwater Freshwater Freshwater and sa ltwater Total (l evel of income) Below median 3 18 15 36 Median 13 28 19 60 Above median 67 125 87 279 Total (type of license) 83 171 121 Table 7.4 contrasted the type of license owned by the particip ant and the level of income. The majority of participants possessed a freshwater licen se; the second largest license group comprised of those who possessed both freshwater and saltwater licenses. The smallest number of participants was those who possessed saltwater licenses only. The largest number of participan ts (125) possessed a freshwat er license and self-reported having above median levels of income. The fewest number of part icipants (3) held a saltwater license and reported having below median levels of income. The VIF value for all variables were negligible (<10), suggesti ng that the models variables have little known pr oblem of multicollinearity. The interpretation of the oddsratio was examined in context to its effect on viewing of fi sh advisory information. This will be discussed in the access to risks information section (pp. 60) of the discussion chapter. 7.3 Field Surveys, direct and participant observations 7.3.1 Field socio-demographics The preliminary questions c onsisted of self-reported soci o-demographic questions. The results from these questions we re summarized in Table 7.3.
Table 7.5: Summary table based on se lf-reported demographic questions Self-reported socio-demographics (N= 77) Variable Categories Number of respondents Ethnicity White 40 (51.9%) Non-white 37(48.1%) Education Some high school/ High school graduate 47 (61.0%) Vocational school 5 (6.5%) Some college/ College graduate 21 (27.3%) Post-graduate/ Other 4 (5.2%) Income Below median 47 (61.0%) Median 16 (20.8%) Above median 11 (14.3%) Did not specify 3 (3.9%) The response rate of the field surveys was 93.9% The total number of participants for the field survey was 77 participants. The major ity of those were Caucasian (51.95%). Nonwhites comprised of those who identified themselves as Hispanic or Latino (19.5%), African American (18.2%), As ians (2.6%), Native Amer ican and Other (both 3.9% respectively).Most of those surveyed were high school graduates 35 (45.5%), many 14 (18.2%) did have some college education. On ly one person had a post-graduate degree. The majority of field participants (61.0%) were below median income (less than $40,000 per year), with the larges t proportion (24.7%) of field su rvey participants earning between $30,001 and $40,000 per year. The fewest respondents (11) produced more than the median level of income (more than $60,000 pe r year). It should be noted that four persons indicated that they were unemploye d and the household income they disclosed was based on the former years income. There was an uneven representation of th e sexes in the field survey. 11 of the 77 participants were women (14.2%). The rema ining 66 were males above the age of 18. Female participation from field survey is si milar to online female participation (12.3%). Participants were also aske d how many persons lived in their household and the ages of any children. The sphere of influence is de picted by figure 7.1 below. The majority of survey participants and persons residing w ith them were adults. The second largest represented group (14.8%) compri sed of children between the ages of 10 and 18, living 51
with the survey participants There were fewer respondents with young children less than three years of age or younger than one year. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% < 1yr. Age 1-3 Age <10 Age 10-18 # of adults <1 10-18 52 Figure 7.1: Sphere of household influe nce including household age dynamics. 7.3.2 Fishing behavior among field participants Self-reported fishing behavior was characterized by answers to quest ions concerning: 1) fishing frequency, 2) if the individual or thei r family consumed the fish caught, 3) what fish were typically consumed and 4) what crit eria was used in selecting the fish eaten. Most fisherfolk (approximately 84%) surveyed stated that recreation was their primary reason for fishing (Figure 7.2). Many of these persons indicated that recreation was their only reason for fishing, as they practiced catch and release. However, some (10.3%) did acknowledge that food was a secondary reason as they ate the fish they caught. There were 12 participants (16%) of the sample that stated they ate the fish. This was because Sphere of household influence Adults 10-18 <10
they believed as it was a free source of prot ein and felt the fish caught was the reward associated with fishing. For example, res pondent 2 surveyed at Rowlett Park stated: I have eaten fish from here (Hillsborough River) for years now. It is free and you have to eat meat to live. Figure 7.3 shows that most participants (51, 66.2%) fished at least once per week. Only one respondent stated that they rarely ever fi shed. Only three persons that were surveyed stated that they fished at least five days per week. In a reconnaissance study, results revealed 93% of sample fished all year. Most of the persons surveyed, 84.4% were Tampa Bay (Hillsborough, Pinellas or Manatee) residents. 0 10 20 30 40 50 60Recreation Food Reason for fishingNumber of respondents Secondary reason Primary reason n=77 Figure 7.2: Rationale for fishing 0 5 10 15 20 25 30FrequencyNumber of participants More than five days/ week between three and four days/ week Once/week one or two times/month few times a year Rarely n=77 Figure 7.3: Fishing freque ncy among participants 53
14% 10% 76% No Sometimes Yes Figure 7.4: Responses from questi on 12: Do you eat the fish caught? Figure 7.4 demonstrates the proportion of participants that indicated that they ate the fish caught. It should be noted that the majority (58, 76%) stated that th ey did eat the fish. Some (8 respondents) indicated th at they were selective about the fish they decided to eat and only ate the fish sometimes. 14% of the samp le indicated that they did not eat the fish at all. All the respondents that did not eat the fish used the terms catch and release. For example, participant number 39, an avid fisherman, who was surveyed when launching his boat at Lowry Park said: I do not eat the fish. I am a proponent of catch and rele ase. I participate in fishing competitions and we all release the fish afterwards. As indicated in figure 7.4, most of the respondents ate the fish they caught. Noting the type of fish surveyed fisherfolk typically caught, consumed and their selection criteria used in determining which fish to consume was important in determ ining risk behavior. Respondents were asked what fish do you catch and eat regularly? The first three fish stated were used for this analysis. 54
Largemouth Bass 16% Gar 7% Catfish 18% Grouper 4% Snapper 3% Perch 9% Red fish 7% red Snapper 2% Shark 6% Sheephead 7% Snook 7% Trout 4% Grunt 2% Ladyfish 2% Mackerel 2% Mullet 4% Largemouth Bass Gar Catfish Grouper Snapper Perch Grunt Ladyfish Mackerel Mullet Red fish red Snapper Shark Sheephead Snook Trout Figure 7.5: Fish identified as being commonly caught and eaten by respondents. This estimation was based on a cumulative count of the first three fish identified as being eaten by each respondent. 55 The fish consumed among the surveyed fish erfolk was quite diverse. Figure 7.5 shows those fish which more than one respondent reported as being regularly consumed. Fish were categorized in the other category only if one respondent indicated they consumed that particular type of fis h. Fish which comprised the o ther category included Cobia, Crappie, Flounder, Florida Po mpano, Walleye, Yellow jack, S tingray and Tuna. Catfish and Largemouth bass were identified as bei ng the most popularly consumed fish among participants. 18% of respondent s indicated they regularly caught and consumed catfish. It should be noted that no catfish species distinction was made among responses. From personal observation, many responde nts believed channel catfish, Ictalurus punctatus was an easy catch and had good meat. Largemouth bass was similarly popular (17%). The consumption of saltwater fish was a lo t more diverse than the freshwater fish consumed. Each identified category had si milar consumption popularity. Redfish or red drum, Sciaenops ocellatus and sheepshead, Archosargus probatocephalus were identified as being consumed and eaten by 7% of su rveyed fisherfolk. Shark showed similar popularity (6%) among the surveyed fishermen. No species distinction was made in shark
56 responses. However, personal observations at Ballast Point did reveal that most of the sharks being caught and kept we re Atlantic Sharpnose Sharks, Rhizoprionodon terraenovae which usually ranged between 12 inches to 24 inches in length. Grunt only accounted for 2% of all fish being identifie d as caught and eaten. All grunt responses were obtained from four surveys with pa rticipants on the Tarpon Springs deep sea excursion. Participant observati ons revealed that these fish were an easy catch and there were no size restrictions on this species. Mo st grunt caught by occupants of the fishing vessel were taken by the fishing guide, Rusty for freezer storage (refer to figure 7.6(A) and 7.6(B)). Rusty was the guide for a chartered fishing boat which launched from the Tarpon Springs sponge docks. Rusty became a key personnel in this study. He was an avid fisherman, guide and boatman and was very knowledgeable of state fishing regulations. He dictated which fish were kept on the fishing excurs ion. He ensured that pe rsons only kept those fish which met fishing regulations, in term s of species and size. After docking, fish caught was later filleted for consumption by Rusty. In an interview with Rusty, he revealed inva luable fish risk safety practices among his clients and his individual consumption risks perceptions. He affirmed that his clients were not overtly concerned about mercury in fish nor did not ask about mercury levels in fish. He discerned that all fish had some leve l of mercury, but some more than others. He said that: Mercury not found in high levels for these fish....deep sea fish. They are found in higher levels in top predators. Indeed, the white grunt and sn apper occupants were catching were of low and moderate consumption risks respectively. He also st ated that he did not consume much high predatory fish, such as king mackerel as they had a lot of mercury.
Rusty also informed me that the Environm ental Protection Agency was now trying to phase out lead sinkers, as they have trace levels of mercury. They wished chartered fishermen in Florida would start using porcelain sinkers. Fishermen in California have already made the transition. (A) (B) Figure 7.6: (A) Fish caught, kept and are stored in ice troughs. (B) The fish was given to occupants upon disembarkation for future cons umption. All the fish in this bunch were grunt, Haemulon plumieri 7.3.3 Self-reported risk behavior among fishing participants After asking participan ts questions relating to fish cons umption and behavior, there were three questions which were asked to determ ine whether people were knowledgeable that mercury was found in fish and that the metallic neurotoxin caused adverse health effects. If participants had claimed they had seen information mentioning that mercury was found in different types of fish, had their indivi dual behavior changed as a result of known contamination to local fish sources. 57
Figure 7.7: Risk understanding and reported behavioral changes Figure 7.7 shows how a breakdown of how many participants were aware that mercury was present in fish. Secondly, how many cha nged their behavior after acquiring the knowledge that mercury is possibly present in the fish they ate. 20.8% (16 respondents) indicated that they did not know what mercury was and seemed genuinely perplexed by the question. 79.2% of the sample respondents (6 1) indicated that th ey were aware that mercury was found in some fish. Of those part icipants who knew mercury was present in fish, only thirteen (16.8%) changed their behavior in an effort to re duce their individual risk. Thirteen participants who indicated th ey changed when they found out mercury was present in Tampa Bay fish, only a couple (6, 7.8%), demonstrated a behavioral change that could be considered an effective mean s of reduced individual consumption risks. 64 respondents (62.2%) indicated that they knew me rcury was present in some of the fish, but did not change their behavi or. Many believed that they di d not eat the fish frequently enough to be concerned with the issue. Respondent 17, who worked in the food processing industry believed that she was not at risk as she did not eat tuna every day. 58
59 Some ate fish of high concern regularly, as they loved the meat. One notable response from Ballast Point: I come out here just to catch shark. It is one of the cleanest places. I eat and catch every chance I got. Some believed that they possessed the knowledge that reduced their individual consumption risks but many others did not possess this knowledge. For example, respondent 18 believed that: You got to cut off the tail of the shark and a lot of blood come out. The mercury coming draining out. Many pe ople dont know how to get rid of the mercury. Several surveyed fishers believed that consump tion of fish from areas with more currents or movement of water were safer than areas with limited circulation, such as Hillsborough River. Respondent 10 at Ballast Point stated that: Different areas have different leve ls of mercury. Some areas are more polluted than others. I prefer Picnic Island, the water is not stagnant. It [Mercury] moves to areas wher e the water is stagnant. Likewise participant 3 believed that mercury deposition and levels of mercury in fish was dependent on local physical conditions. He postulated that: Tampa Bay fish are polluted. I do not eat fish that I catch from here. I only eat the fish that I catch in the Gulf.
60 The majority of the respondents were initially intrigued by the study and wished to know the levels found in the fish. Many showed in creasing concerns and asked if tests were being conducted on the local fish as part of this study and wished to know if it was safe to consume them. At that point in the survey, the surveyor would gi ve the respondent the Hillsborough County Environmental Protec tion Commission Consumption Guide and suggest they follow the guide, pointing out the different risk categories and highlighting the fish they stated they ate. It was empha sized that the guidelines should be followed, especially if there were members of their family within one of the identified high risk groups. The body language and reception of the information was noted. In many instances, the participants were surprised that they were given info rmation at the end of the survey. There was a difference between men and women in the reception of mercury advisories. Some men were not concerned about the health risks associated with fish. Two male respondents said thanks for the info rmation and indicated that they will give their wives or significant other, as they were more concerned with health issues, such as mercury poisoning. Many of the male respondent s think that women need to worry more about consumption risks and shrug at th e possibility of mercury poisoning. 7.3.4 Comparing online and field socio-dem ographic and risk understanding results Two sampling techniques: online random survey and field convenience were utilized in this study. The online survey allowed license holders with email addresses an equal opportunity of being recruited. The field conve nience survey allowed resident fishermen and fisherwomen, who were exempt from licen se possession an opportunity to participate in the survey (appendix 6 outlines license exemptions). In addition, field surveys provided an opportunity to obt ain detailed explanatory data A sample size of 77 was obtained for the field survey, whereas a sample of 375 for the online survey was utilized. This section utilizes descriptive statis tics to compare the sampling techniques.
61 Table 7.6: Summary table for online and field results Variable Category Field (%) Online (%) Ethnicity White 51.9 89.9 Non-white 48.1 10.1 Income Below Median 61.0* 9.6 Median 20.8* 16.0 Above median 14.3* 74.4 Fish consumption (%) 76.0 88.0 Knowledge that mercury could be harmful to human health (%) 79.2 99.9 Self-reported change in behavior (%) 16.8 28.4 This statistic does not add up to 100% as 3 persons did not specify their income. Table 7.6 illustrates summary results from sections 7.2 and 7.3. There were a larger proportion of white participants for the online survey, when compared to the field survey (Table 7.6). Conversely, there was greater representation and diversity of non-whites from the field convenience survey, when compar ed to online survey sampling technique. In addition, there was greater representation of persons with below median and median levels of income from the field convenie nce sample (Table 7. 4).The majority of participants from both sampling techniques c onsumed the fish. There were approximately 20% more online participants reporting that they were knowledgeable that mercury could be harmful to human health. Most participants (79.2%, 358 from both online and field surveys) stated they knew mercury was toxic and present in fish. However, approximately twice the percen tage of online participants indicated that they changed their behaviour in response to viewing of a fish advisory, when compared to the field convenience sample. The majority of online and field respondents that adopted an effective means of effectively reducing their risk did so by reducing their consumption of fish in general or avoided certain types of fish (species or type of production) of fish. For example, field respondent 18 stated: I do not eat much fish anymore. I eat more chicken. Likewise online re spondent 131 wrote:
62 I eat less fish. Few respondents (2) indicated that they ate th e same fish species but chose to consume fish smaller in size. Online respondent commented: Only eat certain species or certain size of species. Younger fish had less mercury Eating smaller sized fish of a concern species is one way individuals would reduce their consumption risks without discrediting benef its from eating a favorite species. Mercury levels are highly correlation to tropic leve ls and weight of fish (Cleckner, 1998; Hammerschmidt and Fitzgerald, 2006). A combination of field and online survey ri sk behavior responses were utilized to determine whether self-reported risk behavior effectively reduced theoretical risks, when mercury risks information was previous ly viewed. When examining responses collectively to question 22: ..has the information influenced your fish consumption? there was approximately 50% dichotomy of amplification and attenuation responses. Figure7.8 shows a few of the verbatim quotes suggesting risk behavior does not mimic the suggested fish guidelines and individuals reported beha vior reflected the acceptance or rejection of consumption ri sks. One online respondent stated that he stayed away from fish which are known to give ciguatera poisoning. While both ciguatera and total mercury in fish biomagnifies up the food chain and tend to be greater in predatory fish such as grouper, ciguatera poisoning incidence cas es are fewer than mercury poisoning incidences (Mebs, 1998). Many persons had many theories as to why their health was not being jeopardized by consumption of local fish. Some of these th eories included: know ledge of the physical environment of their fishing location and th e acquisition of risk information that many fishermen/ fisherwomen were not privy to. Fo r example field respondent 10 believed that
Picnic island wave turbulence and conti nuous cleansing by oceanic currents rid the environment in this particular location of any mercury or waste. Field respondent 34 believed that mercury could be excreted by catf ish and made this part icular fish safer to eat. Other respondents indicat ed that mercury risks c ould be reduced by special preparation of the fish. Field respondent 57 believed that cutting the fin off the shark and allowing the blood to drain reduced the consum ption risks. Likewise field respondent 67 believed that cutting around the fin and bac kbone, where most mercury is stored would reduce risks. None of these theories have scientific evidence and at this time could only be considered myths. 63 Verbatim quotes Figure 7.8: Verbatim quotes used to analyze risk behaviour Survey respondents did not only voice safety concerns for local fish, but also commercially bought fish. Some participants indicated that the geographic location and type of production was critical in deciding if fish in general would be consumed. Online respondent 162 would not eat a ny fish imported from Asia. Presently among American
64 consumers, the media and science can be he ld responsible for stigmatizing Asian seafood. In addition to the stigma associated with seafood from this region, Jacquet and Pauly, (2006) noted that North Americans with highe r levels of education and environmental involvement are influenced by eco-labels and environmental production criteria. Respondent 25 believed that wild salmon was sa fer to eat than farm raised salmon and stated: Now purchase wild salmon over farm raised salmon. Safety views among scientists varied with conc erns to mercury levels in wild and farm raised fish. Wong, (2001) argued that mercury levels in catfi sh, trout and catfish were 40 to 100 times lower than Food and Drug Admini stration safety standards. Whilst other studies such as Debruyn, et al (2006) noted that mercury leve ls in farmed raised salmon and adjacent wild fish stocks were highe r due to higher fecal matter in ponds. These contradictory findings are pub lished, which may leave many consumers confused as to what to purchase and consume. 7.4 Qualitative results derived from interviews with agency officials An informal interview with a Florida Depa rtment of Environmental Protection (FDEP) official relayed that many governmental health and environmental o fficials are concerned with high risk groups mercury exposure through fish consumption, but also believed that the risks of all other individuals who are not categorized within these high risks groups tended to be downplayed or attenuated. In addi tion, the FDEP official relayed that these guides do not consider recent health findings. Epidemiological studies, such as Etherson et al ., (2003) have found a correl ation between in creased consumption of mercury contaminated predatory fish and the increase d incidence of corona ry heart disease among men. There were expressed conc erns that recreational fisher men in Florida are at high
65 risk and should be outlined as a high ri sk group. The quote below was taken verbatim from the Florida Department of Environmental Protection official: Many men eat a lot of fish with high mercury. They think that they are being healthy eating more fish. Fish is recommended by American Heart Association as they are high in omega-3s. Research has shown, many fish high in omega3s tend to have high levels of mercury, such as tuna and mackerel. The informal in terview provided invaluable information and rationale for some of the Florida Fish and Wildlife Conservation Commission maximum size restrictions. An informal interview with an FDEP official re vealed that maximum size restriction was emplaced to reduce consum ption risk to persons eating fish from local freshwater fish sources. For exam ple, in south Florida, only one bass may be 14 inches in total length or longer (refer to section 2.2.1). Cleckner, et al ., (1998) research showed that mercury bioaccumulated within a given individual fish and biomagnified with increasing trophic levels. Figure 7.9 shows the relationship be tween increasing total length of fish and log10 total mercury levels. Although, so me of the Florida Fish and Wildlife Conservation Commission restrictions may have been emplaced to reduce public health consumption, the public is not being in formed as to the reason why these are being enforced. It is imperative that Florida Fish and Wildlife Conservation Commission communicate the reasons for maximum size rest rictions; as opposed to just outlining them, if fish advisories are to be taken seriously.
Figure 7.9: Correlation between total fish length for larg emouth bass (at Loxahatchee) and total log 10 mercury levels, United States Geological Survey, 2003 Verbal and email interactions with Florida De partment of Health and Florida Department of Environmental Protection re vealed risk management concerns as decisions are based on limited historic poisoning incidence and e nvironmental data. The Florida Department of Health could only provide a complete y ears data based on 2008 mercury poisoning reports from registered physicians, although th e database started in 2003. Florida rates for 2008 are shown in figure 7.10. 66
Created by: Trina Halfhide Figure7.10: Mercury poisoning rate (number of persons diagnosed with mercury poisoning per 100,000 persons) using Florida Department of Health 2008 data Figure 7.10 shows the reported mercury poisoning ra te for Florida. It should be noted that all reported cases are within coastal counties. Figure 7.10 demonstrates that Martin County had the highest reported rate of mercury cases. The second highest reported rate was in Broward County. Pinellas and Manatee Counties within the defined study area had reported mercury rates of 0.212620 and (0.2126210.254932) per 100,000 respectively. The results from this study revealed many key findings related to community awareness of mercury in locally caught fish in Tampa Bay. The discussion re-examines the results in context of the research questions and relates key findings to pr evious studies. In addition, limitations of this study and policy implicati ons would be discussed in the chapter 8. 67
68 Chapter 8: Discussion 8.1 Introduction This study examined the community awarene ss of mercury in fish utilizing a mixed methods approach. The discussion will re-examine the results in context with the research questions: 1) Are mercury consumption risks associated with fish being communicated effectively by environmental policy makers (state and local government)? 2) Do all fisherfolk have equal access to, and understanding of risk consumption information that allows for informed decision making? and, 3) Does mercury risk information change behavior among fisherfolk? The mixed methods approach used in this st udy not only facilitated different segments of the fishing population to be sampled, but al lowed the results from these two survey sampling methods to be contrasted and increased the validity of the study. Online surveys permitted an examination of generalized patt erns that influenced viewing of mercury related risk information whilst, field surveys f acilitated in-depth analysis of risks to be determined. In addition, key pers onnel interviews with agency officials were used to gain an understanding of agency officials sen timents of current state mercury laws. 8.2 Research question 1: Are mercury cons umption risks associated with fish being communicated effectively by environmental po licy makers (state and local government)? The agencies outlined in this study: Florid a Department of Environmental Protection (FDEP), Florida Department of Health (FDOH), Florida Fish and Wildlife Conservation Commission (FFWCC) and Florida Department of Agriculture and Consumer Services. The aforementioned agencies had very limite d visible collaborative policies, programs
69 and guidelines as it relates to mercury issues. Florida Department of Health responsibilities include determining the potenti al for adverse human health effects from consuming the local fish and issuing fi sh consumption advisories, whilst the Environmental Protection Commission of Hillsborough County implements federal programs at the local level and also issues advisories. The guidelines outlined in local Hillsborough County Environmental Protection Co mmission advisories are in most cases consistent with the State advisories. Both agencies have highlighted women of childbearing age, pregnant women and young ch ildren as being high risk groups in their consumption advisories. Although the guideline s in the fish advisories tend to be consistent between agencies, the Florida Department of Agriculture and Consumer Services (FDACS) recommendations (figure 2.7 ) did conflict with the other agencies guidelines. FDACS reco mmended consuming predatory fis h, such as shark and swordfish twice a week to maintain a healthy heart. This difference in information may be a reflection of their interest groups they repres ent. The Florida Department of Agriculture and Consumer Services (FDACS) represent the interest of state agriculturalists and commercial fishery industries. Large disparities in the conveyance of risk information suggest there is the need to have one unify ing committee to address mercury issues. In addition, interviews with agency officials (s ection 7.4) emphasized the need for agencies to collaborate and devise a fish advisory th at includes recent risk findings, such as men with elevated blood mercury levels associated with high consumption of fish was correlated to high incidence of co ronary heart disease (Etherson et al ., 2003). In addition to the interagency divide, there was an apparent disconnect between agency officials and lay persons. This paradigm lock may be due to the belief that a higher priority should be placed on vulnerable gr oups: women and children. The agencies do not seem to be overly concerned with higher in cidences of coronary heart disease among men associated with elevated blood mercury levels (Etherson et al 2003). The attenuation of mens risk may exist because men innately tend to have higher body masses than women and children and can therefore be exposed to higher concentrations without experiencing any adverse effects from which a safe level, or reference dose is derived. Risk assessment
70 methodology consists of several concepts and may not be easily understood by professionals not involved in risk assessment. Safe levels for the different types of fish are determined using a known body weight of 70kg and considers that any given individual would be eating a portion of fish (six ounces). However, most persons customarily eat more than six ounces of fish for a given meal. The Florida Department of Health consumption guideline ou tlines the portion size of fish that should be consumed, however it is unknown how many people unders tand this information and follow the stipulated portion sizes. Figure 7.10 showed that all the reported rates of mercury we re in coastal counties. In addition, this figure also showed that there were cases in the de fined Tampa Bay study area. These results are consistent with Karouna-Renier et. al (2008) study, authors showed that coastal populati ons consumed more fish th an inland populations which correlated with higher blood mercury levels Although the rates are usually higher among coastal population, the rates depicted in figure 7.10 ar e deceptive and only show confirmed cases of mercury poisoning. Howe ver, mercury poisoning associated with contaminated fish presents an insidious risk as symptoms are similar to other neurological diseases. As a result, many cases could be mi sdiagnosed as other diseases such as early Parkinsons disease and the cases are underrepo rted. It is estimate d that 85,000 children in United States are born annually at risks for neurological de fects associated with fetal exposure to mercury (Anderson et al ., 2004). These cases are not typically documented by the Florida Department of Health as a mercury poisoning case. According to Slovic (2000), decision makers within agencies are forced to construct simplified models to deal with risks as ther e are cognitive limitations associated with the science of risks. In this study, the certainty of risks to the general fishing population is unknown. Unknown outcomes lead to the rationa lization of the construction of basic models, such as those being utilized by state agencies to deal with mercury issues. The lack of reliable historical mercury poisoning incidence data reflects the failure of modern scientific institutions and agencies to addr ess technological risks with rigor reduces the
71 respect of the public which they are trying to protect. Often times, risks are rationalized and politicized within the agency and the scientific object ive is lost. Burger (2005) study produced similar results and concluded that fish advisories are often politicized and agency officials tend to base their decisions on guiding principles rather than the science behind the risk. This makes uniformity of messages among governmental entities difficult. In addition, in the same study it was noted that there tende d to be a shift of responsibility from agencies to individuals in reducing individual health risks. Results from this study also suggested that the fishing community may have heard that mercury was present in fish but was not inform ed of the details of the consumption risks or the rationale for consumption freshw ater maximum size restrictions. Active educational outreach is crucial if recreationa l fishermen and Tampa Bay residents are to be knowledgeable of the mercury consumption risks presented to them. Publishing consumption guides is the first progressive step in community outreach; however effective communication of complex scientific risks would not be achieved in a short period and should be part of long-term goal of agencies. Building lasting relationships with the fishing community re quires continuous liaison on the part of agencies. Education should be the focal point in any human heal th or environmental plan. Many may argue that if fishermen and fisherwomen becam e more aware and followed consumption guidelines, there will be lowe r stock levels of fish at lo wer trophic levels with lower risks, such as Blue Gill and Redear Sunfish, as persons would overfish the edible, low risks fish. However, state officials believe that most fishermen ar e not overly concerned of the risks presented to them as risks ha ve not been communicated effectively and therefore ecological inst ability is not a major concern in the immediate future. Ecological assessments could determine if stock levels of game or popular fish are dwindling and revisions of to existing minimum size restrictio ns are necessary for outlined game fish.
72 8.3 Research question 2: Do all fisherfolk have equal access to, and understanding of risk consumption information that al lows for informed decision making? Awareness to mercury consumption risks info rmation was assessed using field and online surveys. As previously mentioned, online su rveys were conducted to assess a larger segment of the local fishing population and de termine variables that contributed to the viewing of mercury related risks information. Whilst field surveys were utilized to gather robust qualitative data from survey participan ts, they also allowed explanatory risks information to be delivered. Although, the overall response rate for the on line study was 5.58% when considering the total recruited population size of 7334, the resp onse rate did vary between individual databases. There was a zero response rate for databases with military sportsmen and lifetime licensees (persons 65 and older). The low response rate among military sportsmen may be because they are still fulfill ing their serving duties. The low response rate among the lifetime licensees was similar to Deutskens et al ., 2004 which showed 1.8% response rate among respondents ages 65 to 74. In addition, th is studys results showed an overrepresentation of younger and higher education persons. Similarly, this study had approximately 63.2% of respondents between the ages of 18 to 50 and 77.6% of participants had receive d at least some college level of education. Cook et al ., (2000) argued that the representativeness of samples was more important than the response rate. The online survey was distributed to the en tire population of licensees with email addresses with the defined study area of Hillsborough, Manatee and Pinellas Counties to allow all fishermen equal opportunity to pa rticipate in the survey. There was an overrepresentation of individuals which were Caucasian (89.9%), possessed an above median level of income (74.3%) and men (87.7 %). While most of the participants were Caucasian, this proportion tended to be consistent with the census of (81.3%). Likewise, the proportion of women (12.3%) was similar to the proportion of licenses sold to
73 women. A future study could utili ze a stratifying sampling me thod and try to obtain more individuals within those group iden tified as being underrepresented. The odds ratio from the binary l ogistic regression results is used to interpret how a given variable of significance contributed to the viewing of mercury related risk information. Results revealed that the use of internet s ources for public health information and having a freshwater license significantly influen ced viewing of mercury consumption risks related information (p<0.10). In addition, sex, ab ove median levels of income and type of license did influence the viewing of consumption risk information (p<0.05). These variables contribution to view ing of mercury related risk information are discussed in greater detail below. 8.3.1 Internet media sources The result from the binary l ogistic regression showed that the odds of viewing mercury consumption risk information are 1.748 times mo re when utilizing internet sources of public health information, when other variables are held constant. The internet to date can no longer be considered a luxury. Internet has become an important tool for communication, knowledge and earning a living in modern society. Most agencies relay natural and technological risk information or da ta on the web in real time or live. Florida Department of Health publishes its state advi sories online. Results from this study were consistent with Hesse et al ., 2005 study, which stated that approximately 63.6% of persons in the United States with internet access have utilized internet sources of information for themselves or others within a 12 month period. It should be noted that Chakraborty and Bosman (2005) results showed income ownership in equality in personal computer had been continuously declining; although the Southern United States where this study was based experienced the greatest di gital divide in all of the United States. Unequal or limited access to internet sources could imply that there is unequal access to public health and risk information. Although, et hnicity was not a sign ificant variable in influencing viewing of fish a dvisory information in this study, it should be noted that
74 Chakraborty and Bosman (2005) results di d show that income inequalities among computer household have decreased ra pidly among whites between 1994 and 2001; however, comparative results were not seen with African Americans. If risk information is being increasingly available on the internet and governmental agencies issuing these advisories reduce the use of other sour ces, disenfranchised or less economically privileged individuals may become more vulnerable to health risks, such as mercury poisoning due to the indiscriminate cons umption of contaminated fish. Risks could be amplified or attenuated by inte rnet media sources. The ability to update information using live news feeds and reports allows audio and virtual messages to be seen by the majority of modern society within seconds of an incident happening. Stigma and experiences become embedded before all facts are gathered. Creating experiences for viewers may be counterproductive for risk intuition as peopl e become desensitized to claimed dangers. 8.3.2 Type of license Type of license did influence viewing of mercury information (p<0.05). The odds of viewing mercury consumption informa tion are 0.328 times less when possessing a freshwater only license and possession of a saltwater license is used as the reference variable. Likewise, the odds of viewing mercury consumption information are 0.318 times less when possessing both freshwater and saltwater license and possession of a saltwater license is used as the reference variable. Results indicated that people were most interested in capturing saltwater species, ev en if they were at a freshwater site. Most of the game fish sought after tended to be saltwater game fish. In addition, many avid fishermen with saltwater licenses tend to read fishing magazine and actively seek information about the fish they are captur ing. Many avid fishermen tend to acknowledge that predatory fish have more mercury than the species in the lower trophic levels and always highlight saltwater top predator game species, such as tuna and neglect to mention any freshwater ones. This may be due to saltw ater fish being regular ly sold and familiar
75 to the general public, including survey part icipants, compounded by there being more effective communication of saltw ater fish consumption risk by agencies, including Food and Drug Administration (FDA). Re searchers established that mercury levels in saltwater fish tend to be higher than in freshwater fish (Kannan et al ., 1998; Cleckner et al ., 1998). Viewing of risk information may be consistent with the fish that can be purchased at their grocery store. More people wish to consume saltwater, so they seek or view information that are consistent with their consumption in terests. Results suggesting that people were more knowledgeable of consumption risks associ ated with saltwater fi sh were consistent with other risk communication st udies. Burger, 2005 and Verger et al ., 2007 have established that participants were more aw are of the high risk fish which have been processed such as, canned tuna. 8.3.3 Sex Results showed that although there were mo re male than female respondents (12.3%), which Z proportion is not significantly different (p< 0.01) when comparing proportions of regional licenses purchase s (American Sportsfishing Association, 2005). However, the American Sportsfishing Association estimated that 29% of licensed fishers in Florida are women. In many studies there are notably mo re male fishermen. Burger (2004) notes there were significantly more men that fished than women in a similar study done in Central New Jersey. Fishing tends to be a male dominated activity (Anderson, et al ., 2004). It should be noted that even though fish ing tends to be a male dominated activity, there was an inherent sampling error when more than one licensed fisher used the same email contact, as in the case of husband and wife at the same residence. In was observed from sensory viewing that most of the em ails would correspond with the mans name. Survey monkey, the survey provide would not distribute the ema il twice, although there were two different persons with the same listed email contact. The odds of viewing mercury consumption risk information are 0.390 times less when a female licensee is considered and all other va riables are held constant. This outcome is
76 consistent with results from similar studies. Anderson et al (2004) notes awareness of advisories varied significantly by sex, women being less aware of local fish advisories and specifics relating to the levels of mercury to the trophic levels of fish. The authors related their results to the premise that s port-fish advisory communication programs have traditionally targeted licensed anglers, who were predominantly white males. It should be noted that although women were generally less aware about mercury advisories, all women (46) surveyed knew that mercury in fi sh was detrimental to human health. Results from z-test for proportions test revealed vi ewing of risk information between women of reproductive age (18-50) and women of non-re productive (> 50) was not significant (zscore =1.042, p>0.01). These result s suggest that there are no differences in viewing or access to information within the female licensee participants. Younger women of reproductive age, a target audience group fo r fish advisories were not viewing significantly different than women of other age cohorts. While, these empirical results suggested that women were viewing the info rmation less, no conclusions could be drawn to determine whether patterns of viewing mimicked the perceptions and understanding of the actual risk within this subgroup. 8.3.4 Income The odds of viewing consumption advisories were 1.990 times greater when the licensee has an income above the national median (40,000-60,000 per annum) and all other variables are held constant (p<0.05). 74.3% of those surveyed licensees fell within this category. In this study, the number and type of license varied with income level (table 7.4). The largest proportion of participants po ssessed above median levels of income for all types of licenses (freshwate r only, saltwater only and bot h saltwater and freshwater). The largest number of individuals possessed a freshwater license. One would expect that income would be linked to greater access to media resources, especia lly internet sources and this may have played a role in increased viewing of fish advisories. However, above median levels of income and the internet as a risk information source were not highly correlated (p>0.05).
77 Although the quantitative binary logistic regression results were able to provide an indication of factors influencing viewing of advisory information, the qualitative data results stemming from open-ended questi ons provided explanatory data into understanding the dreaded consumption risk of eating mercury contaminated fish. There were a few cases in which participants believed that the fish was safe to eat as there were no advisory signs at the stat e parks or publicly assessed fishing areas. One respondent stated adamantly that no advisories were issued for Hillsborough River as there were no signs posted at any of the fishing points. In addition to inequity in information access, there were several misconceptions. Misconcepti ons could be considered as rationalized dissonance without any theore tical risks reduction. Most pa rticipants (79.2%, 358 from both online and field surveys) stated they knew mercury was toxic and present in fish. Many understood that mercury was a contaminan t and had ill effects on the environment and human health in general. However, more in-depth field surveys revealed that although participants claimed th ey knew mercury was present in fish and had deleterious health effects; most did not understand the fate of mercury in the environment or within an individual fish. 8.4 Research question 3: Does mercury risk information change behavior among fisherfolk? The risks associated with consumption of me rcury contaminated fish are considered a low probability, high consequence event, acco rding to Slovics (2000) classification. The effects associated with such events or poi soning events tend to be long lasting. This section examines how participants perceived consumption risks associated with mercury in fish and if behavior changed in response to risk perceptions. Most participants (79.2%, 358 from both online and field surveys) stat ed they knew mercury was toxic and present in fish. Many understood that mercury was a contaminant and had ill effects on the environment and human health in general. Th e majority of respondents that demonstrated an effective means of reducing consumption ri sk did so by reducing their consumption of
78 fish in general or avoided certain types of fi sh (species or type of production) of fish. However, many in-depth field surveys reveal ed that although partic ipants claimed they knew mercury was present in fish and had de leterious health effects; most did not understand the fate of mercury in the envi ronment or within an individual fish. Furthermore personal communica tion with respondents reveal ed several misconceptions as to how an individual c ould reduce ones consumption ri sks. Misconceptions could be considered as rationalized dissonance without any theoretical risk s reduction. Section 7.3.4 outlined the many theories people believed as to why an individuals health was not being jeopardized by consumption of local fis h. Some of these theories mentioned in the results included: knowledge of the physical environment of their fishing location, the acquisition of risk informati on that many fishermen/ fisher women were not privy to and confidence in future government health sy stems. A risk communication study by Sustein, 2002 examining the perception of radon contamin ation revealed that many people held many inaccurate beliefs. Participants beliefs and the lack of knowledge pertaining to radon exposure made the problem seem more severe and unsolvable than the reality of the risk. A full examination of all self reported changes in behavior revealed an approximately 50% dichotomy between risk attenuation and amplification responses. These results may be due to the nature of both fishing as a recreational activity and mercury as a dreaded risk. Recreation was the primary reason for fishing for most and any risk associated with the voluntary activity itself c ould be deemed as negligible. However, mercury could be deemed as an involuntary, unknown dreaded risk, which not all is known to science (refer to figure 3.1). Any given indi vidual must weigh the benefi ts and risks of their own situation. For example, individuals who refused to eat fish because they are knowledgeable that mercury is found in fish would also reduce ones health benefits derived from consuming fish, such as lo wer cardiovascular disease risks. Some respondents perceived that the dreaded risk s associated with mercury poisoning far outweighed the benefits. One female survey respondent indicated that her sister, a marathon runner experienced mercury poisoning and refused to eat any fish caught. She
79 practiced catch and release (section 7.3.4). Ho wever, some believed that the benefits associated with consumption of fish far outweighed any dreaded risks associated with eating contaminated fish. For exam ple, online respondent 162 stated: No have not changed consumption. Health benefit far outweighed the concerns. There were two online responses which demons trated that particip ants had knowledge of higher risk individuals and a change of behavior occurred to protect such individuals family members. For example, online respondent 130 replied: Eat less fish during wifes pregnancy. Careful about feeding children fish. Both Florida Department of Health and Hillsborough County Environmental Protection Commission have highlighted young children, wo men of child bearing age and pregnant women as higher risk groups. The consumpti on guide produced by Florida Department of Health recommended different consumpti on frequencies for the vulnerable risk individuals and another for a ll other individuals. Although only two responses suggested that individuals knew that women and children were higher risk groups, Burger et al ., (1999) results demonstrated that pregnant women were less likely to know about consumption advisories than non-pregnant women. It should be noted that while sustainability was mainly examined in the co ntext of sustaining health, this study also examined how risk perception may affect sustai nability of the fishery resource (refer to Section 3.4). If suddenly, people became increas ingly aware of mercury in fish, they may drastically change their behavior. Furthe rmore, the fishing community may possibly consume fish in a lower trophic level or overfis h the smaller sized predatory fish, such as Swordfish. In observance w ith the Millennium Development Goal 7: ensuring environmental sustainability, if behavior s of fishermen/ fisherwomen do change drastically, a study should be conducted to assess the ecological stability of game fish. An
80 ecological shift may mean that rivers and la kes should be restocked or there should be stricter permits guidelines. The results of this study have revealed critical flaw s in existing mercury risk communication management policies and programs. Results have suggested that messages vary according to the in terest of the individual agency and there is a need for future interagency collaborations. Policy im plications are discussed in more detail in section 8.6 (policy implications). 8.5 Limitations and future scope for the study There are a few limitations of this study and these should be addressed. Firstly, this study did not take any biological samples and ther efore no definitive correlation could be made between participants fish consumption a nd actual levels of mercury exposed to. However, other studies such as Mahaffey et al (2008) and Harada (1998) have determined that there is a correlation between the contaminated fish consumed and blood mercury levels and mercury levels in hair, respectively. One of the limitations associated with ethnograp hic research is that many persons tend to report the most current behavior or concerns and do not necessarily express their deepest concerns in a single interaction. For example, the species identified as being consumed was linked to the time of the year or th e fish they had caught that day. Six field participants surveyed duri ng the months of November 2008 and March 2009 identified sheepshead as being regularly caught and ea ten. However, sheepshead is a migratory species and is found in the Tampa Bay Re gion during the winter months. Likewise, participants in the Tarpon Springs area identif ied deep sea fish that they were catching the same day as being caught and consumed on a regular basis. In addition, most ethnographic studies including this one would have restrictions based on the sampling method or the surveying technique utilized. Five publicly accessed fishing areas were
81 selected based on their popularity among local fishermen and identification of the site being within a mercury impaired site (USE PA, 2008). In addition, these study sites may not be representative of the entire Tampa Bay subsistence and recreational fishermen population. In addition, online surveys were used because they were more amenable to the other segments of population, such as affl uent individuals with boats than paper-andpencil surveys. Slovic (2000) notes that wh ite males who usually have higher incomes tend to perceive risks less th an their female and non-white counterparts. A combination of paper-and-pencil or field surveys and on line surveys were utilized to obtain a wider range of participants with di fferent socio-demograpics. Using selection criteria: licens ee must be 18 years or older, correctly geocoded with an address within Tampa Bay (Hillsborough, Manatee or Pinellas counties) and having a registered email address in the database ma y suggest that there may be some inherent biases in the sampling procedure used. The la tter criterion is proba bly the most exclusive of the three criteria. In additi on, the binary logistic regressi on only used an entry if the respondent completed the survey; entries without all invested independent parameters were omitted from the binary logistic regr ession analyses. Although these stipulations may be considered excessive by some, the st udy still revealed stat istically significant findings relating to acces s of mercury consumption risk re lated information. In addition, the quantitative binary logi stic regression model determined which discrete sociodemographic variables influenced fish adviso ry viewing. This model did not consider the role political and cultural f actors played. However, Weber and Hsee (2009) results demonstrated cross cultural differences between participants was a stronger determinant for risk perception than income and occupa tion. Only one question examined differences in food preparation. Future re search could examine crosscultural differences in the perception of mercury in fish.
82 8.6 Policy implications Results revealed that risk communication by agencies did not facilitate optimal risk reduction among recreational and subsisten ce fishermen in Tampa Bay. Although, the development of a fish advisory is supposed ly a joint collaboration, an analysis of agencies literature revealed that recommendations and fish consumption guidelines did not relay the same information. Florida Depa rtment of Consumer and Health Services (FDACS) suggested eating fish such as, shark and swordfish at le ast twice a week to maintain a healthy heart. However, Florida De partment of Health suggested that persons in high risks groups (women of childbear ing age and young children) should not eat swordfish. This conflict in information conveyance may be indicative of the specific interest groups they represent. Attenuation of mercury risks and conflicting information may result in the lay public not trusting ag encies, which are responsible for ensuring public health is maintained. An independe nt ombudsman, who is preferably a riskcommunication specialist, should overlook the fish advisory pr ocess to ensure the risk messages being conveyed reflect the science and not guiding principl e. Training is the risk communication arena is necessary to develop a tailored program. Figure 8.1 shows a conceptual mediated risk communication m odel, which considers all groups which contribute to the risk comm unication process. Sustein, 2002 argued that too much information might even make people less informed as people have a limited capacity to process information, such as complex inci dence rates such as 1 per 100,000 may be difficult to conceptualize. Although, it is the re sponsibility of agencies to ensure people understand the insidious risk and heed a dvisories, people facing pervasive risk information may experience information overload and may cause persons in the public to treat the information as no information. In addition, Truckner (2009) results revealed that 93% of the health care professionals indicated that th ey did not distribute humaninduced environmental degradation information, such as mercury contamination to their patients, although 75% of respondents indicated that there was an unfulfilled need for education concerning advers e health impacts associat ed with human-induced environmental degradation.
83 Results suggest that people are genera lly not overly concerned with mercury contamination in fish, especially freshwat er fish. Explanatory data from surveys demonstrated that persons were more aware of mercury in saltwater fish. Although, saltwater fish tend to pose more of a potenti al risk, the public shoul d be aware of local freshwater fish that are contaminated. US EPA recommends limiting consumption of local freshwater fish that are not under state advi sories. However, most people consult their state advisories over federal ones as thes e agencies are more locally active. The freshwater guide issued by the Florida Fish and Wildlife Conservation Commission outlines maximum size freshwater restrictions for freshwater fish that are notorious for high mercury levels. However, the guide does not state the reason for the maximum size restrictions. In addition, familiari ty of species of fish may play an important role in how the person perceives the consumption risk. Ma ny responses suggested that participants favored saltwater fish to freshwater fish, liv ed in the Tampa Bay area for a long period and had favorite fishing spots and had not s een any reasons for c oncern. More in-depth research examining cultural dynamic and peopl es background beliefs would be helpful in determining the optimal risk communica tion program agencies should adopt when tackling fish advisories. In this study, results demonstrated that women viewed fish advisories less than their male counterparts. In addition, women of reproducti ve age, a target group for fish advisories had not viewed fish advisories signifi cantly more than all other age cohorts. Approximately, 4.7 million (16%) women of ch ildbearing age nationwide are estimated to have potentially unsafe blood mercury leve ls and about 630,000 new borns are at risks of neurological and developmental health problems. Indiscriminate consumption of mercury contaminated fish presents an insi dious risk to recrea tional and subsistence fishermen in Tampa Bay. More proactive measures should be adopted to protect vulnerable populations. In meeting long-term objectives of this study, fish advisory warning signs will be placed along fishing s ites along the Hillsborough River to increase local fishermens awareness of mercury levels in the different species of fish. These sign
84 will mimic the Environmental Protection Commission consumption guidelines (figure 2.5).
POLITICAL INSTITUTIONS AND POLICY MAKERS 85 LAY PUBLIC OPPOSITION PARTIES CORPERATIONS THE MEDIA SYSTEM SCIENTIFIC AND EXPERT COMMUNITIES CAMPAINING GROUPS AND SOCIAL MOVEMENT lobbying PR expert information and recovery Lobbying and advice direct communication suppor t PR PR direct co mmunication support support risk experiences and o p inions representation of risk events and issues Figure 8.1: Mediated risk Co mmunication Model, adapted from Durant and Lindsey (1999,pp.5)
86 Chapter 9: Concluding Statements With increasing industrialization and ener gy consumption but limited environmental remediation to reduce environmental and heal th burdens associated with mercury, fish advisories remain the primary means of reducing human exposure to mercury. Results from the binary logistic regression suggest ed that the following parameters were statistically significant when considering viewin g of fish advisories: sex, internet sources of information, above median levels of income, sources of public health information and type of license. In addition, open-ended respons es suggested that people were expecting to see signs at fishing sites, if there was a local advisory in effect. There were no signs posted at the publicly accessed fishing areas, whic h gave some a false sense of security in fish consumption safety. Currently, mercury c onsumption risks associated with fish are not being communicated and understood in such a manner that ensures equal access to the decision making process to maintain healt h. Both field and online surveys established that although people stated that they kne w mercury was found in fish, many did not understand the fate of mercury in the envir onment or fish. There were many instances where individuals attenuated mercury consum ption risks, whilst others amplified it. Agencies and health care providers do unders tand that the information is not being relayed to all members of public. Yet, still there is passive educational outreach using limited media.
87 References Agency for Toxic Substances and Disease Registry. (1999). Toxicological Profile for Mercury Association Argawal, A., Gibson, C.C., (Eds.). Communities and the Environment: Ethnicity Gender, and the State in Community-Based C onservation, New Brunswick: Rutgers University Press. Anderson, H.A, Hanrahan, L.P., Smith, A., Dr aheim, L., Kranarek, M., Olsen, J., The role of sportfish consumption adviso ries in mercury risk communication: a 1998-1999 12-state survey of women 18-45, Environmental Research 95, 315324 Aronson, E., (1997), Back to the Future: Retr ospective Review of Leon Festinger's" A Theory of Cognitive Dissonance, The American Journal of Psychology 110 (1), pp. 127-137 Atkeson, T.D., Mercury in Florida's Environment Florida Department of Environmental Protection, Retrieved from http://www.dep.state.fl.us/labs /mercury/docs/flmercury.htm Attwell, R. I. G. (1970). Some effects of Lake Kariba on the eco logy of a floodplain of the Mid-Zambezi Valley of Rhodesia. Biological Conservation, 2, 189-196. Baxter, J. & J. Eyles. (1997). Evaluating Qualitative Research in Social Geography: Establishing 'Rigour' in Interview Analysis. Transactions of the Institute of British Geographers, 22, 505-525. Beehler, G., P, B. McGuinness, M. & J. Vena, E. (2001). Polluted fish, sources of knowledge, and the perception of risk: Cont extualizing African American anglers' sport fishing practices. Human Organization, 60 288. Bergan, T., Gallardo, L., Rodhe, H., Merc ury in the global troposphere: a threedimensional moddel study, Atmospheric Environment, 33, 1575-1585
88 Bernard, S., A. Enayati, L. Redwood, H. Roger & T. Binstock (2001). Autism: a novel form of mercury poisoning. Medical Hypotheses, 56 462-471. Bischi, G. I. & F. Lamantia (2007) Harvesting dynamics in protected and unprotected areas. Journal of Economic Behavior & Organization, 62, 348-370. Burger, J. (2000a) Consumption advisories and compliance: The fi shing public and the deamplification of risk. Journal of Environmenta l Planning and Management, 43, 471. Burger, J. (2000b) Gender Differences in Meal Patterns: Role of Self-Caught Fish and Wild Game in Meat and Fish Diets. Environmental Research, 83, 140-149. Burger, J.S., Megan Fitzgerald, Suzie Flores and Caron & Chess (2008). Fish consumption: efficacy among fishermen of a brochure developed for pregnant women. Journal of Risk Research, 11 pp.891-904. Burger, J., K. K. Pflugh, L. Lurig, L. A. Von Hagen & S. Von Hagen (1999). Fishing in Urban New Jersey: Ethnicity Affects Information Sources, Perception, and Compliance. Risk Analysis, 19, 217-229. Burger, J., Sanchez, J. & Gochfeld, M. (1998). Fishing, Consumption, and Risk Perception in Fisherfolk al ong an East Coast Estuary. Environmental Research, 77, 25-35. apek, S.M., The "Environmental Justice" Frame: A Conceptual Discussion and an Application, (1993), Social Problems 40 (1), 5-23 Carbone, F. & Accordi, G.(2000). The Indian Ocean Coast of Somalia. Marine Pollution Bulletin, 41, 141-159. Chapman, L. & Chan, H. M. (2000). The In fluence of Nutrition on Methyl Mercury Intoxication. Environmental Health Perspectives, 108, 29-56. Chakraborty, J., Bosman, M.M., Measuring the Digital Divide in the United States: Race, Income and Personal Computer Ownership (2005). The Professional Geographer 57(3), 395410 Clarkson, T. W. Magos, L. & Myers, G.J ., The Toxicology of MercuryCurrent Exposures and Clinical Manifestations (2003). The New England Journal of Medicine, 349, 17311737.
89 Cleckner, L. B., Garrison, P. J., Hurley, J ., P., Olson, M.,L., Krabbenhort, D., P (1998) Trophic transfer of methyl mercury in the northern Florida Everglades. Biogeochemistry, 40, 347-361. Collings, S. E., M. S. Johnson & R. T. Leah (1996). Metal contam ination of anglercaught fish from the Mersey Estuary. Marine Environmental Research, 41 281297. Connealy, L. E., M.D. (2006). The Mad Ha tter Syndrome: mercury and biological toxicity Cooke, S. J. & L. U. Sneddon (2007). Animal welfare perspectives on recreational angling. Applied Animal Behaviour Science, 104, 176-198. Counter, S. A. & L. H. Buchanan (2004). Mercury exposure in children: a review. Toxicology and Applied Pharmacology, 198, 209-230. Cutter, S. (1994). Environmental Risks and H azards. (Ed.). New Jersey: Prentice Hall. Daskalakis, K. D. & T. P. O'Connor (1995). Distribution of chemical concentrations in US coastal and estuarine sediment. Marine Environmental Research, 40, 381-398. Eby, L. A., W. J. Roach, L. B. Crowder & J. A. Stanford (2006). Effects of stocking-up freshwater food webs. Trends in Ecology & Evolution, 21, 576-584. Fitzgerald, W. F. & T. W. Clarkson (1991).Mercury and Monomethylmercury: Present and Future Concerns. Environmental Health Perspectives, 96 159-166. Florida Fish and Wildlife Conservation Commission. 2008. Florida Fish Regulation Freshwater Edit. ,2008-2009. Flynn, J., Slovic, P. & Kunreuther, Risk Media and Stigma: Understanding Public Challenges to Modern Science and T echnology. London: Earthscan Publications Ltd. Freese, C. H. 1997. Harvesting Wild Species. The John Hopkins University Press Gilmour, C. C. & G. S. Riedel (200 0). A Survey of Size-Specific Mercury Concentrations in Game Fish from Maryland Fresh and Estuarine Waters. Archives of Environmental Contamination and Toxicology, 39 53-59. Haines, T. A., T. W. May, R. T. Finlay son & S. E. Mierzyko wski (2003). Factors Affecting Food Chain Transfer of Mercury in the Vicinity of the Nyanza Site,
90 Sudbury River, Massachusetts. Environmental Monito ring and Assessment, 86, 211-232. Hammerschmidt, C. & W. Fitzgerald (2006) Bioaccumulation and Trophic Transfer of Methylmercury in Long Island Sound. Archives of Environmental Contamination and Toxicology, 51, 416-424. Harada, M., J. Nakanishi, S. Konuma, K. Ohno, T. Kimura, H. Yamaguchi, K. Tsuruta, T. Kizaki, T. Ookawara & H. Ohno (1998). The Present Mercury Contents of Scalp Hair and Clinical Symptoms in Inhabitants of the Minamata Area. Environmental Research, 77, 160-164. Hesse, B.W., Nelson, D.E., Kreps, G.L., Cr oyle, R.T., Arora, N.K., Rimer, B.K. Viswanth, K., Trust and Sources of Hea lth Information: The Impact of the Internet and Its Implications for Health Care Providers: Findings From the First Health Information National Trends Survey, Archives of Internal Medicine Vol. 165 (22), 2618-2624 Holt, C. A. & R. M. Peterman (2008) Uncertainties in population dynamics and outcomes of regulations in sockeye salmon (Oncorhync hus nerka) fisheries: implications for management. Canadian Journal of Fisheries & Aquatic Sciences, 65, 1459-1474. Howard, J.A. Halfhide, T.; Thomas, K.D.; Omis ca, E.; Michael, R.; Stuart, A.; Akiwumi, F.; Trotz, M. A. Mercury in the Hills borough River in Tampa, Florida: Fish consumption advisories and water, sediment and fish loadings. Submission planned for Environmental Monitoring and Assessment Jaffe, D., Prestbo, E., Swartzendruber, P., Weiss-Penzias, P., Kato, s., Takami, A., Hatakeyama, S., kajii, Y., (2005) Export of atmospheric mercury from Asia, Atmospheric Environment 39 (17), 30293038 Kannan, K. & J. Falandysz (1998).Speciation an d Concentrations of Mercury in Certain Coastal Marine Sediments. Water, Air, & Soil Pollution, 103, pp.129-136. Kar, S.B., Alcalay, R., Alex, S. (Eds.). 2001. Health Communication: A Multicultural Perspective London: Age Publications, Inc. Karouna-Renier, N. K., K. Ranga Rao, J. J. Lanza, S. D. Rivers, P. A. Wilson, D. K. Hodges, K. E. Levine & G. T. Ross (2008) Mercury levels and fish consumption
91 practices in women of child-bearin g age in the Florida Panhandle. Environmental Research, In Press, Corrected Proof 7. Kasperson, R.E., Renn, O., Slovic, P., Brown, H. S., Emel, J., Goble, R., Kasperson, J.X., Ratick, S. (1988). Social Amplication of Risk: A Conceptual Framework, Risk Analysis Vol. (8), No.2, pp.177-187 Latern Corperation,(2002) Trends of Anth ropogenic Mercury Mass Flows and Emissions in Florida, Clayton, MO: Authors, Husar, J.D. & Husar R.B. Lash, S., Szerszynski, B, Wynne, B. (Eds.) (1996). Risk, Environment and Modernity. London: Sage Publications. Laukkanen, M. (2003). Cooperative and non-coope rative harvesting in a stochastic sequential fishery. Journal of Environmental Economics and Management, 45, 454-473. Lawson, N. M. & R. P. Mason (1998). A ccumulation of Mercury in Estuarine Food Chains. Biogeochemistry, 40, 235-247. Macauley, J. M., J. K. Summers, V. D. Engle & L. C. Harwell (2002). The Ecological Condition of South Florida Estuaries. Environmental Monitoring and Assessment, 75, 253-269. Mahaffey, K. R., R.P. Clickner, R.A.Jeffr ies (2008). Adult Womens Blood Mercury Concentrations Vary Regionally in USA: Association with Patterns of Fish Consumption (NHANES 1999-2004). Environmental Health Perspectives, Online. Mann, B. Q. (1995). Quantification of illicit fish harvesting in the Lake St Lucia Game Reserve, South Africa. Biological Conservation, 74, 107-113. Mathias Kondolf, G. (1994). Environmental pl anning in regulation and management of instream gravel mining in California. Landscape and Urban Planning, 29 185199. McDowell, M. A., Dillon, C. F. Osterloh, J., Bolger, P. M., Pellizzari, E., Fernando, R. R. Oca, M. D. Schober, S. E., Sinks, T. Jones, R. L.& Mahaffey, K. R. (2004). Hair Mercury Levels in U.S. Child ren and Women of Childbearing Age:
92 Reference Range Da ta from NHANES 1999-2000. Environmental Health Perspectives, 112, 1165-1171. McPhee, D., (2008), Fisheries Management in Australia Sydney: The Federation Press Mebs, D., (1998), Occurrence and Sequest ration of Toxins in Food Chains, Toxicon 36 (11), 1519-1522 Mihelcic, J.R., Crittenden, J.C., Small, M.J., Shonnard, D.R., Hokanson, D.R., Zhang, Q., Chen, H., Sorby, S.A., James, V.U., Su therland, J.W., Schnoor, J.L., (2003), Sustainability Science and Engineering: The Emergence of a New Metadiscipline, Envir. Sci. Techno ., 37, 5314-5324 Moyle, P. B. (1995). Conservation of native fr eshwater fishes in the Mediterranean-type climate of California, USA: A review. Biological Conservation, 72, 271-279. National Institute of Environmental Health Sc iences, United States Department of Health and Human Services. (2008). Adult Wo men's Blood Mercury Concentrations Vary Regionally in USA: Association with Patterns of Fish Consumption (NHANES 1999-2004), 1-44. doi: 10.1289/ehp.11674 (available at http://dx.doi.org/) Oses-Eraso, N., Udina F.,& Viladrich-Gr au, M. (2008).Environmental versus HumanInduced Scarcity in the Commons: Do They Trigger the Same Response? Environmental & Resource Economics, 40, 529-550. Pandit, A. K. (1991). Conservation of wildlife resources in wetland ecosystems of Kashmir, India. Journal of Environmental Management, 33, 143-154. Paustenbach, D. J. (2000).The Practice of Exposure Assessment: A State-of-the-Art Review. Journal of Toxicology & Envi ronmental Health: Part B, 3, 179-291. Popkin, B.M, Guilkey, D. K., Haines, P. S., (1989), Food Consumption Changes of Adult Women between 1977 and 1985, American Journal of Agricultural Economics 71 (3), 949-959 Pichichero, M. E., E. Cernichiari, J. Lopreiato & J. Tr eanor (2002). Mercury concentrations and metabolism in infants receiving vaccines containing thiomersal: a descriptive study. The Lancet, 360, 1737-1741.
93 Reader, S. & Zandergeren, P. (2002). A Geodemographic Analysis of Marine Recreational Fishing on the West Coast of Florida Papers and Proceedings of Applied Geography Conferences, Vol. 25,138156 Riley D. M., Newby, C.A., LealAlmeraz, T.O. (2006). Incorporating Ethnographic Methods in Multidisciplinary Appr oaches to Risk Assessment and Communication: Cultural and Religious Uses of Mercur y in Latino and Caribbean Communities. Risk Analysis, 26, 1205-1221. Ruseski, G. (1998). International Fish Wars : The Strategic Roles for Fleet Licensing and Effort Subsidies. Journal of Environmental Economics and Management, 36, 7088. Shimshack, Jay, Michael B.Ward, and Timothy K.M. Beatty, (2007).M ercury Advisories: Information, Education, and Fish Consumption, Journal of Environmental Economics and Management, 53 (2), 158-179. Sigler, W. F., Sigler, J.W. (1984). Recreat ional Fisheries: Management, Theory and Application. Sheehan, K., (2001), E-mail Survey Response Rates: A Review, J ournal of ComputerMediated Communication, 6 (2), Retrieved from: http://jcmc.indiana.edu/vol6/issue2/sheehan.html Shrivastava, P., (1995), Democr atic control of technological risk in developing countries, Ecological Economics, 14, 195-208 Shorrock, A.F. (1975). The AgeWealth Relationship: A Cross-Section and Cohort Analysis, A Review of Econom ics and Statistics 57 (2), 155-163. Sjberg, J.,(2000), Factors in Risk Per ception, Risk Analysis 20(1), 1-11 Sjberg, J.,(2004), Explaining Indi vidual Risk Perception: The Case for Nuclear Waste, Risk Management: An International Journal, 6(1), 51-64 Slovic, P., (2000). Perception of Ris k, London: Earthscan Publications Ltd. United Nations Economic Commission for Europe (1990) Bergen Mini sterial Declaration on Sustainable Development Bergen, UNECE
94 Trisak, J., (2005). Applying game theory to analyze the influence of biological characteristics on fishers cooperation infisheries co-management, Fisheries Research Vol. 75, 164-174 Truckner, R.T., (2009), Health Care Provider Beliefs Concerning the Adverse Health Effects of Environmental and Ecosystem Degradation, Wilderness and Environmental Medicine 20, 199-211 Urban Environmental Management, Retrieved from: http://www.gdrc.org/uem/footprints /index.html, last accessed 8/1/08 United States Environmental Protection Agency, Retrieved from: http://oaspub.epa.gov/waters/env iro.control?p_list_id=FL-1443D last accessed on 3/13/08, last updated on 3/13/08 United States Environmental Protection Agen cy, Executive Order: Federal Actions to address environmental justice in minority population and low income populations, Retrieved from: http://www.epa .gov/Region2/ej/exec_order_12898.pdf United Stated Geological Survey, Ener gy Resources Program, Retrieved from: http://energy.er.usgs.gov/ United States Geological Survey, 2009,[Graph ill ustration Correlation between total fish length for largemouth bass (at Loxahatchee) and total log 10 mercury levels, United States Geological Survey, 2003], South Florida Information Access, Retrieved from: http://sofia.usgs.gov/projects/int _geochem_foodweb/isotope_geer03abs.html United States Department of Health a nd Human Services, and U.S Environmental Protection Agency, Retrieved from: http://www.cfsan.fda.gov/~frf/sea-mehg.html United States Department of Health and Human Services, (1999). P ublic Health Service Agency for Toxic Substances and Disease Registry: Toxicological Profile for Mercury, Government Printing Office. Got Mercury?: A project of Turtle Isla nd Restoration Network, Retrieved from: www.gotmercury.org/
96 Appendix 1: Glossary of terms Age connotes the number of years that the nu mber of years that the respondent has experienced life. Community : In this study, the concept of community refers to those that have the shared common interest of fishing in the Tampa Bay Region. Dependent variable : Viewing of fishing advisories will be used as the de pendent variable in the proposed study. Ethnicity in this study connotes indivi duals that share cultural, linguistics, and behavioral traits and includes racial characteristics. The participant is asked how they would identify themselves. Household income refers to the collective moneta ry consumption opportunity those persons in a household in a ye ar. Income groups were catego rized for the purpose of this study. The lowest income gr oup defined was less than te n thousand dollars and the largest income group defined was greater than one hundred thousand dollars. Independent variables : Independent variables include ethnicity, household income, sex, age, occupation and highest education leve l achieved. Ethnicity, sex, occupation and highest educational level achieved are qual itative and discrete categories, whereas household income and age will be used as quantitative but discrete as they are categorized into categories in the semi-structured survey. Occupation refers to persons main source of inco me and livelihood and is an open ended question in the survey. This criterion woul d consider part-time and full-time positions.. Occupational vulnerability was characterized by the recognition of the stated occupation within an industry considered as possi bly having a mercury exposure by Florida
97 Department of Health. Occupational vulnera bility did not necessitate an obvious exposure risk to that individua l. Persons within a recognized higher risk industry, such as those working in the medical field may know about mercury from on the job training and their use of instruments which contain mecury. Response rate : The percent of persons that that part icipated in your survey divided by the total number of respondents recruited for the survey. Sex in this study refers to the external biological expression of men and women. It is anticipated that this variable would be used to determine if risk perception differs by sex. Tampa Bay resident : Anyone having a residence or zip code in Hillsborough, Pinellas or Manatee Counties.
98 Appendix 2: List of geocoding databases created The following is a list of the va rious address databases created: 01Hillsborough resident 5-year freshwater licenses 02Hillsborough full resident freshwater licenses 03Hunting and freshwater combination licenses 04Hillsborough freshwater, saltwater and hunting combination licenses 05Hillsborough gold sportsman licenses 06Hillsborough resident sportsman licenses 07Manatee captain licenses 08Manatee 5-year saltwater licenses 09Manatee full-resident saltwater licenses 10Manatee 5-year freshwater resident licenses 11Manatee full-resident freshwater licenses 12Manatee freshwater and sa ltwater resident licenses 13Manatee Life Sportsmen 14Hillsborough Freshwater and saltwater resident licenses 15Hillsborough chartered licenses 16Hillsborough lifetime saltwater licenses 17Hillsborough 5-year saltwater 18Not applicable 19Pinellas Resident saltwater licenses 20Pinellas over 65 sportsman licenses 21Pinellas Resident saltwater licenses 22Pinellas lifetime saltwater licenses 23Pinellas lifetime sportsman 24Pinellas military gold sportsman 25Pinellas 5 year resident 26Pinellas fresh salt
99 27Pinellas resident gold sportsmen 28Pinellas charter
100 Appendix 3: Email invitation sent to licen sees in Hillsborough, Manatee and Pinellas Counties Hello Florida Anglers, Fishing serves as both a popular recreational activity and an important source of food for many of us who utilize Florida's lakes, rivers, gulf, and bays. The preservation of a healthy fishery is critical to these interests. Research at the University of South Florida is focusing on methods to help provide a healthy and sustainable fishing resource for recreational anglers. My name is Trina Halfhide and I am writing to ask for your participation in this nonprofit research project. I am University of South Flor ida (USF) graduate student examining fishing habits and consumption pa tterns of locally caught fish in our Tampa Bay community. The attached survey should ta ke no longer than 10 minutes of your time. You can stop participating at any time. Y our responses will be used for statistical analyses and no individual data will be publis hed. If you have any questions or concerns about this study or the manner in which this survey was conducted, please feel free to contact Fenda A. Akiwumi, PhD. She can be contacted via email at firstname.lastname@example.org or telephone at (813)974-6887. Please double click on link to start survey: http://www.surveymonkey.com/s.aspx?sm=lORBguq1mLK4SHdc_2fTk_2fyg_3d_3d Thank you in advance for participating in this study and taking an additional step to make Tampa Bay's resources more sustainable.
Appendix4: Semistructured su rvey used in this study Park and internet semi-structured surv ey among Tampa Bay recreational fishers Date: Sex: Questions: 1. What is your age? less than 18 18 to 25, 26 to 35, 36 to 50, 51 to 65, over 65 If the subject is less than18, indicate that the survey is for persons older than 18 years of age and end survey as indicated on page 1. 2. How would you identify your ethnicity? Caucasian, African American, Hispanic, Native Indian, East Indian, Mixed, Asian, 3. Do you live in the Tampa Bay area (Hillsbo rough, Pinellas or Manatee Counties)? Yes, No 4. What is your most visited or favorite fishing site? ______________ 5. What is your occupation? ________________________ 6. How many years have you been in this occupation?__________ 7. What is your highest level of formal education? some elementary, through 8th grade, some high school, high school graduate, some college, college graduate, vocational (trade) school, post-graduate degree, Other (specify)______________ 8. What is your household income in US dollars? less 10,000, 10,001 to 20,000, 20,001 to 30,000, 30,001 to 40,000, 40,001 to 60,000, 60,000 to 100,000, more than 100,000 9. How many persons live in your household including yourself? ______ 10. How many persons are in the following age groups? ( ) Adults (greater than18years), ( ) Children between ag es 10-18 years, ( )less than10 years, (_)between1 to 3 years, (_)less than 1year 101
102 11. How often do you fish? More than 5 days/ week, 2 to 4 days/ week, 1 day/ week, a few times a year, rarely, never 12. Do you and your family eat the fish caught? Yes, No, Sometimes 13. What are some of the fish you catch and eat regularly? _________________ _________________ _________________ 14. How do you decide which fish to keep? Size, Type of fish, Keep all, other, Specify ______________ 15. What are your reasons for fishing? food, recreation, income (money) 16. How often do you eat fish in general? 3 times or more/ week, 1to 2 times/ week, 1or 2 times/ month, few times a year, rarely 17. How do you prepare the fish (choose more than one option where applicable)? fried, ceviche, baked, grilled, stew 18. Do you eat all parts of the fish including skin, cartilage? Yes, No 19. What are your main sources of public health information? Radio, Television, Internet, Municipal Bulletins, Employer, Doctor, Other(specify)______ 20. Are you aware that some fish contain mercur y levels that may be harmful to some people? Yes, No 21. Have you seen any information mentioning th at mercury is found in certain types of fish? Yes, No 22. If yes to the previous question, has the information influenced your fish consumption? Yes, How? __________________________________, No This brings us to the end of our survey. Thank you for your participation. If you wish to obtain more information about mercury in fish, you can refer to Florida Department of Health at: http://www.doh.state.fl.us/environment/commu nity/fishconsumpti onadvisories/fish_eat ing_guide_eng.pdf
103 Appendix 5: Semi-structured inte rview questions and topics 1. To gain a historical perspectiv e of mercury policy within Florida. 2. To determine if there are any differences between USEPA and FDEP guidelines in what is considered 'impaired waters' 3. To determine if there are any hot spots. Are these synonymous with impaired waters? 4. To determine if the Tampa Bay is consid ered a hot spot, when taking into account water/ sediment environment? Or consumption issues? 5. To gain a greater unders tanding of consumption issues as it relates to policy. What is the role of each agency in th e policy process, including FDEP? 6. To obtain whatever data that would strengthen my study.
104 Appendix 6: Exemptions to purchasing a license These exemptions were taken verbatim from the Florida Fish and Wildlife Conservation Commission (FFWCC) website. Retrieved from: http://myfwc.com/License/LicPe rmit_RecreationalHF.htm#exempt You do not need a hunting, freshwater fishing or saltwater fishing license or a *Florida waterfowl, turkey, snook, spiny lobster, ar chery season, crossbow season, muzzle-loading season, or management area permit if... You are a child under 16 years of age. (Also exempt from federal duck stamp requirements.) You are a Florida resident age 65 or olde r possessing proof of age and residency or possessing a Resident Senior Citizen Hun ting and Fishing Certific ate. Residents age 65 or older may obtain, at no cost, complimen tary hunting and fishing certificates from county tax collectors' offices. You hunt or freshwater fish in your county of residence on your homestead or the homestead of your spouse or minor child, or if you are a minor child hunting or freshwater fishing on the ho mestead of your parent. You are a Florida resident certified as totally and permanently disabled and you possess a Florida Resident Disabled Pers on Hunting and Fishing Certificate. You are a resident who is a member of the Armed Forces of the United States, you are not stationed in this state, and you are home on leave for 30 days or less, upon submission of orders. Effective Aug. 1, 2009 You are a resident who is fishing with live or natural bait, using poles or lines that are not equipped with a fishing-line-retrieval mechanism, and you are fishing for noncommercial purpos es in your home county. However, you must have a valid fishing license to fish by any method in a fish management area. These exemptions do not appl y for the federal duck stamp. You do not need a freshw ater fishing license if...
105 You have been accepted as a client for de velopmental disabilities services by the Department of Children and Family Services The department must furnish proof to such clients. You are a resident who is fishing with live or natural bait, using poles or lines that are not equipped with a fish ing-line-retrieval mechanis m, and you are fishing for noncommercial purposes in your home county. However, you must have a valid fishing license to fish by any method in a fish management area. You are fishing in a fish pond of 20 acres or less which is loca ted entirely within the private property of its owner. A fish pond is a man-made pond constructed for the primary purpose of fishing, entirely within th e property lines of th e owner and with no surface water connection to public waters. You are fishing in a fish pond of 20 acres or more, whose owner has purchased a fish pond license at a fee of $3 per surface acre. You possess a Resident Freshwater Commercial Fish ing License. You are fishing in the St. Mary's Rive r or Lake Seminole (but not including tributary creeks in Florida) and have a valid Georgia fishing license. You are freshwater fishing during Free Fi shing Weekend (the first weekend of April). You do not need a saltwater fishing license or a snook or spiny lobster *permit if... You have been accepted as a client for de velopmental services by the Department of Children and Family Services. The departme nt must furnish proof to such clients. You fish from a for-hire vessel (guide, char ter, party boat) that has a valid vessel license. You fish from a vessel, the operator of wh ich has a valid vessel license issued in the name of the operator of the vessel. You fish for recreational purposes from a pi er with a valid pier saltwater fishing license. You have a valid saltwater products license.
106 You fish for mullet in fresh water and you have a valid Florida freshwater fishing license. You are a resident who is saltwater fishi ng from land or a structure fixed to land who has been determined eligible for th e food stamp, temporary cash assistance, or Medicaid Program by the Department of Ch ildren and Family Services (DCFS). Proof of identification and a benefit issuance or program identification card issued by DCFS or the Agency for Health Care Administra tion must be on your person when fishing. NOTE: Effective Aug. 1, 2009 Residents a nd nonresidents fishing for a saltwater species (other than mullet in fresh water) fr om land or from a structure fixed to the land are required to purchase a saltwater fishing license. These exemptions do not apply to tarpon tags.