Development of Social Indicators for Fishing Communities of the

Development of Social Indicators for Fishing Communities of
the Southeast: Measures of Dependence, Vulnerability,
Resilience, and Gentrification
NOAA/NMFS Award Number NA08NMF4270412 (#103)
FINAL REPORT
Lincoln Center, Suite 740
5401 West Kennedy Blvd.
Tampa, Florida 33609-2447
May 2010
This Final Report was prepared by the Gulf & South Atlantic Fisheries Foundation, Inc. under award
number NA08NMF4270412 from the National Oceanic and Atmospheric Administration, U.S.
Department of Commerce. The statements, findings, conclusions, and recommendations are those of the
author(s) and do not necessarily reflect the views of the National Oceanic and Atmospheric
Administration or the Department of Commerce.
Title:
Development of Social Indicators for Fishing Communities of the
Southeast: Measures of Dependence, Vulnerability, Resilience and
Gentrification
Authors:
Gulf & South Atlantic Fisheries Foundation, Inc.
Lincoln Center, Suite 740
5401 W. Kennedy Blvd.
Tampa, Florida 33609-2447
Frank C. Helies, Program Director
Judy L. Jamison, Executive Director
Dr. Steve Jacob, York College of Pennsylvania
Dr. Priscilla Weeks, Houston Advanced Research Council
Dr. Benjamin Blount, SocioEcological Informatics
Award No:
NA08NMF4270412 (Foundation #103)
Project Period:
June 2008 – May 2010 (Amended)
Table of Contents
PAGE
I. Title Page
1
II. Abstract
6
III. Executive Summary
6
IV. Purpose
10
A. Description of the Problem
10
B. Objectives of the Project
11
V. Approach
11
A. Description of the Work Performed
11
1. Vulnerability and Resiliency
(A) The Relationship of Vulnerability and Resiliency to Well-Being
(B) The Time Element of Vulnerability and Resilience
(C) Sustainability, Vulnerability, and Resilience
(D) Measures of Social Vulnerability and Resiliency
(E) Measures of Economic Vulnerability and Resilience
1
11
12
12
13
14
14
(F) Measures of Ecosystem / Natural Environment Vulnerability and Resiliency
(G) Indicators of Social Disruption
2. Dependence and Reliance
15
16
17
(A) Relationship of Resource Dependence on Vulnerability, Resilience, and
Gentrification
(B) Fishing Community Dependency and Social Impact Assessment
(C) Dependence and Dominance in the Local Community
(D) Dependence versus Reliance
(E) Potential Indicators of Fishing Dependence
(1) Economic Dependence
(2) Social Dependence
17
18
19
19
20
20
21
3. Gentrification and Well-Being
(A) Etiology of Gentrification
(B) Urban versus Rural Gentrification
(C) Forms of Rural and Coastal Gentrification
(D) In-authenticity of Coastal Gentrification
(E) Potential Indicators of Coastal Gentrification
(1) Urban Sprawl Indicators
(2) Natural Resources Migration
(3) In-Migration of Retirees
21
22
23
23
25
25
25
26
26
4. Triangulation: Ground-Truthing
27
B. Project Management/Work Performed
27
1. Planning Meetings
(A) Social Indicators Workshop
(B) Social Indicators Workgroup Meeting
(C) Final Planning Session
27
27
30
31
2. Methods and Data
(A) Social Indicators
(1) Social Indicator Indices Development Strategy
(2) Social Indicator Indices Components and Internal Reliability
(3) Community Selection and Inclusion in the Data Set
(4) Data Set Characteristics and Sources
(a) Issues with Confidentiality
(B) Ground-Truthing
(1) Ethnographic Field Research
(a) Field Research Plans and Sites
(b) Ethnographic Interview Protocol
(c) Field Research Methods and Data Collection
(i) Galveston and Galveston Bay Communities
(d) Number of Interviews
(e) Coding Procedures
31
31
31
32
33
33
33
34
34
34
38
38
38
39
40
2
(f) Features of Data Analysis
(2) Compilation of Historical and Contextual Background
3. Project Personnel
40
41
41
VI. Findings
43
A. Actual Accomplishments and Findings
43
1. Social Indicators Indices
(A) Urban Sprawl Index
(B) Natural Resource Migration Index
(C) Retirement Migration Index
(D) Population Composition Vulnerability/Resiliency Index
(E) Poverty Index
(F) Housing Characteristics Vulnerability/Resiliency Index
(G) Labor Force Structure Index
(H) Natural and Technological Disaster Risk Index
(I) Housing Disruptions Index
(J) Personal Disruption Index
(K) Commercial Fishing Reliance Index
(L) Recreational Fishing Reliance Index
(M) Social Fishing Dependence Index
(N) Local Fish Stock Sustainability Index (L-FSSI) and Quotient Indicators
(1) Creation of the L-FSSI and Examples of Interpretation
43
43
45
46
47
48
49
50
51
52
53
54
55
56
57
59
2. Ground-Truthing
(A) Interviews and Coding
(1) Master Sheet and Codes
(2) Explanation of Codes
(3) Coding Results – Raw Scores
(4) Coding Results – Rate Measures
(5) Summary: Code Scores
(6) Keywords, Community Characteristics: Vulnerability and Resilience
(7) Additional Information from the Interviews
(B) Contextual Research
(C) Summary: Ground-Truthing Results
61
61
61
64
65
66
70
71
77
78
84
B. Significant Problems
86
C. Need for Additional Work
86
VII. Evaluation
87
A. Attainment of Goals
87
3
1. Evaluation Strategy
(A) Description of the Social Indicators Process
(B) Description of the Ethnographic Process
(C) Differing Processes with a Converging Reality
(D) Interrater Reliability
(E) Coding Issues for the Secondary Data Indicators
(F) Interrater Agreement Results
87
87
88
88
89
90
91
2. Overall Evaluation of Agreement
94
3. Dissemination of results
94
VIII. Recommendations
95
IX. References Cited
96
Appendix A: Semi-Structured Interview Protocol
103
Appendix B: Historical Sketches of the Coastal Communities
106
Appendix C: Gentrification: Communities
113
Appendix D: Gentrification: Slated Development Projects
120
Appendix E: Impacts of Hurricane Ike
125
Appendix F: Recreational and Commercial Fishing Infrastructure
138
Lists of Figures
Figure 1:
Figure 2:
Figure 3:
Figure 4:
Figure 5:
Figure 6:
Figure 7:
Figure 8:
Figure 9:
Figure 10:
Figure 11:
Figure 12:
Figure 13:
Figure 14:
Map of Galveston, Matagorda, and San Antonio Bays
Map of the Galveston Bay Geographical Entities
Map of San Antonio and Matagorda Bays
Cattle Drive across Redfish Bar, Galveston Bay
Shrimp Ready to be Packed in Barrels on the Galveston Dock, 1920s
Early Twentieth Century Shoreline of Port Lavaca, Texas
Path of Hurricane Ike
Damage in San Leon from Hurricane Ike
Damage to San Leon due to Hurricane Ike
Damage to Texas City Dike from Hurricane Ike
Damage to a Residential Area in San Leon from Hurricane Ike
Changes in Access sites since 1990 – Southern Research Sites
Changes in Access sites since 1990 – Northern Region
(does not include hurricane damage)
Locations of Kemah and Seabrook Docks and Fish Houses
4
34
36
37
107
109
110
126
127
128
129
136
139
140
141
Lists of Tables
Table 1:
Table 2:
Table 3:
Table 4:
Table 5:
Table 6:
Table 7:
Table 8:
Table 9:
Table 10:
Table 11:
Table 12:
Table 13:
Table 14:
Table 15:
Table 16:
Table 17:
Table 18:
Table 19:
Table 20:
Table 21:
Table 22:
Table 23:
Table 24:
Table 25:
Table 26:
Table 27:
Table 28:
Table 29:
Table 30:
Table 31:
Table 32:
Table 33:
Table 34:
Table 35:
Table 36:
Table 37:
Table 38:
Table 39:
Factors That Comprise Community Vulnerability and Resiliency
Factors That Comprise Social Vulnerability and Resiliency
Factors That Comprise Economic Vulnerability and Resiliency
Factors That Comprise Ecosystem/Natural Resource Vulnerability and
Resiliency
Factors That Comprise Social Disruption
Factors that Comprise Forms of Fishing Dependence in Communities
Potential Measures of Commercial, Recreational, and Non-Consumptive
Economic Dependence
Potential Indicators of Social Dependence
Factors that Comprise Forms of Coastal Gentrification in Communities
Indicators of Urban Sprawl
Indicators of Natural Amenities Migration
Indicators of In-Migration of Retirees
Number of Interviews Completed in the Coastal Communities
The Urban Sprawl Index
The Natural Resource Migration Index
The Retirement Migration Index
The Population Composition Vulnerability/Resiliency Index
The Poverty Index
The Housing Characteristics Vulnerability/Resiliency Index
The Labor Force Structure Index
The Natural and Technological Disaster Risk Index
The Housing Disruptions Index
The Personal Disruption Index
The Commercial Fishing Reliance Index
The Recreational Fishing Reliance Index
The Social Fishing Dependence Index
The Local Fish Stock Sustainability Index (L-FSSI) and Quotient
Indicators
Community Landings Data with Gulf of Mexico Landings Data
References by Keywords to Code Topics by Individuals and Community
Rate Measures of Topic References by Individuals and Community
Comparison of Rank of FISHERIES Section Totals by Community
Comparison of Rank of INDIVIDUALS Section Totals by Community
Topics/Keywords that Indicate FISHERIES Vulnerability
Topics/Keywords that Indicate FISHERIES Resilience
Topics/Keywords that Indicate INDIVIDUALS Vulnerability
Topics/Keywords that Indicate INDIVIDUALS Resilience
Summary Scores of Community Vulnerability and Resilience
Summary Scores and Community Rank for Vulnerability and Resilience
Community Economic and Social Dependence on Commercial and
Recreational Fishing
5
13
14
15
15
16
19
20
21
25
26
26
27
40
44
45
46
47
48
49
50
51
52
54
55
56
57
58
60
62
66
70
71
72
74
74
75
77
77
79
Table 40:
Table 41:
Table 42:
Table 43:
Table 44:
Table 45:
Summary of Commercial, Recreational, and Social Dependence
by Community
Community Socioeconomic Vulnerability
Levels of Gentrification in the Nine Coastal Communities
Commercial, Recreational, and Social Dependence
Quantitative Gentrification Indices
Vulnerability/Resilience Indices
81
81
83
92
93
94
II. ABSTRACT
The purpose of this research was to identify potential threats to the commercial and recreational
fishing communities in the South Atlantic and Gulf of Mexico. A workshop was convened to
develop a protocol for (1) the construction of social indicators applicable to coastal fishing
communities, and (2) the evaluation of eleven selected communities through field research.
These concerns were operationalized into a community-level information system utilizing
secondary data as indicators. To assess the external validity of the indicators, field research was
conducted in the selected communities on the Texas coast. The results of the field research were
used to ground-truth the results of the secondary data and to modify and correct where necessary.
The resultant information system will be made available to concerned parties in the public,
private, and governmental sectors to be used in preparing for threats as well as assessing impacts
of proposed fisheries management strategies on communities. Such work is of vital importance
as it will allow the communities that are dependent upon fishing to assess a variety of threats and
vulnerabilities that are both social and ecological. In addition, this information system will
enhance fisheries managers ability to comply with National Standard 8 of the Magnuson Stevens
Act and will mesh well with an Ecosystems Management approach.
III. EXECUTIVE SUMMARY
The problem addressed by the project is the development of reliable methods to provide social
indicators of resilience and vulnerability in relation to sustainability and well-being of fishing
communities. The concepts of resilience, vulnerability, sustainability, and well-being are
relatively new in their consideration of coastal zone management. They have proven useful in
application to coastal zone hazards, especially to disasters in relation to extreme weather events,
hurricanes in particular. They have also been applied to marine fisheries at national and regional
levels but much less to local and community levels. The goal of this research project was to
develop reliable methods to extend those concepts to coastal communities. The development of
a system for measurement of the concepts is a much needed complement to the recent increase in
rapid social impact assessment of coastal communities. While rapid social impact assessments
are needed to produce reliable community profiles, they are not designed to produce needed
measures of how communities fare in times of change, especially highly stressful change.
Community profiles are important, but they are only the initial step in a series of data needs to
assess the impacts of actions or cumulative impacts of regulation and other forces. The
construction of indicators and indices of vulnerability, resilience, sustainability, and well-being
6
will add an important set of assessment measures to the social impact literature for fishery
management.
A number of objectives were developed to guide the research efforts toward identification of
social indicators and indices for coastal communities. The overall objective was to identify
potential threats to the commercial and recreational fishing communities in the South Atlantic
and Gulf of Mexico. A workshop was convened to develop a protocol for (1) the construction of
social indicators applicable to coastal fishing communities, and (2) the evaluation of eleven
selected communities through field research. These concerns were operationalized into a
community-level information system utilizing secondary data as indicators. To assess the
external validity of the indicators, field research was conducted in the selected communities on
the Texas coast. The results of the field research were used to ground-truth the results of the
secondary data and to modify and correct where necessary.
To help meet the overall objective, a number of specific objectives were established.
1. To convene a workshop to identify data and appropriate social indicators for construction
of social impact indices for fishing communities.
2. To develop a set of indices that will assist in the measurement of social impacts at the
community level for fishing communities of the Gulf and South Atlantic.
3. To place those indices into an information system to be linked to other indices of coastal
relevant information, i.e., measures of habitat sensitivity and/or coastal hazards.
4. To conduct an in-depth study of dependence, resilience, vulnerability, and sustainability
within selected communities, using large, secondary data sets and ground-truthing the results
ethnographically.
The workshop was held at the Houston Advanced Research Center on June 15-17, 2008, and it
was attended by regional, state, national, and international researchers and government officials
knowledgeable about coastal issues relevant to the identification and development of social
indicators. The construction of the list of topics followed literature reviews, PowerPoint
presentations, and discussions at the workshop. Consideration was also given to the NOAA
NMFS document, “Guidance for Social Impact Assessment” (2001), which includes five
categories of social factors identified for use in assessing management impacts. These groupings
are: 1) demographics for the community; 2) cultural data related to the fishery; 3) social structure
and institution impacts; 4) cultural data related to the community; and 5) historical and current
participation in the fishery. Although useful as general guidelines, the groupings do not all lend
themselves to operationalization. Some of these five groupings are more easily operationalized
with secondary data using social indicators. These include topics like population size, number of
residents living in poverty, and numerous other factors identified at the workshop and pursued in
7
the research. Identification and development of the indicators constituted the second and third
objectives of the project.
Four sets of factors were identified that comprise community vulnerability and resilience: social;
economic; ecosystem; and social disruptions. Each of these was factored into secondary
categories. Social factors were seen as based on population composition, poverty, and housing
characteristics. Economic factors were seen within economic structure. Ecosystem factors
included natural disasters, technological disasters, and regulatory impacts. Social disruption was
seen in terms of housing, economic, and personal disruptions. Each of these secondary
categories was further sub-divided into data categories.
Consideration was also given to natural resource dependence and well-being. The focus was on
the relationships of dependence with vulnerability, resilience, and gentrification, and relevant
factors were identified for fishing dependence. Fishing dependence was viewed in terms of:
economic factors, including commercial fishing, recreational fishing, and non-consumptive value
dependence; and social factors, including institutions, cultural heritage, and landscape/built
environment. Socioeconomic factors included a cumulative ranking of the economic and social
factors. Gentrification factors included urban sprawl, people attracted to natural amenities, and
in-migration of retirees. Each of the sub-categories was further sub-divided to establish factors
that could be operationalized.
The data set in this research was compiled from six separate data sources. The primary source
for population and housing information was the U.S. Bureau of the Census 1990 and 2000
Decennial Census, Summary Tape File Three. The main source for information about fishery
landings, permits, and value was provided as a custom database by NOAA Fisheries Southeast
Regional Office and Southeast Fisheries Science Center. Very current population estimates and
real estate values were downloaded from the website “City-Data.com.” The data for natural
disaster risks were downloaded from “Moving.com.” The data for technological disasters were
downloaded from the Environmental Protection Agency’s (EPA) website from the
“EnviroMapper” store front. Lastly, data for marinas and related businesses were downloaded
from the 2002 Economic Census and the U.S. Census Bureau’s “American FactFinder” web
page.
Three steps were taken to develop the social indicator indices. First, correlation coefficients
were examined to find underlying patterns of variation. Second, the variables that were most
highly intercorrelated and reflected the range of ideas of interest were placed in a principal
components analysis, where these variables were determined to be reliable indices. Third, the
variables were standardized and weighted for their effects in the model. Index factor scores were
used. The factor loadings are a rough indication of correlation of the domain concept’s latent
structure to the single variable. Items that are most important in an index receive a higher
weighting than a less important item. The factor scores were standardized with a mean of zero
8
and the scores reflecting standard deviations from that mean. Scales were subsequently tested
for internal consistency by using Armor's (1974) theta reliability for factor scales.
There were three groupings of communities primarily based on estuaries that were selected for
study. In the Gulf of Mexico, on or near Galveston Bay, were the communities of Seabrook, San
Leon, Galveston, Texas City, and Bacliff. In the Gulf of Mexico, on or near San Antonio Bay,
were the communities of Port Lavaca, Sea Drift, Port O’Connor, and Palacios on Matagorda
Bay. In the South Atlantic were the communities of Little River and Mount Pleasant. These
eleven communities were located in or adjacent to the following counties in Texas: Brazoria,
Calhoun, Chambers, Galveston, Harris, Jackson, Matagorda, and Victoria. In South Carolina
the communities were located in or adjacent to Charleston, Georgetown, and Horry counties.
Because eleven communities would not provide sufficient variation in the data for reliable index
development it was decided to include all communities in the county and adjacent counties. This
resulted in a data set with 122 different communities and provided sufficient variation for index
development. The results of the community rankings on each variable are reported, however,
only for the eleven communities as identified.
The research design of the project called for ground-truthing of the characterization of the
selected communities from the large, secondary data sets. The ground-truthing part of the
project was designed to independently test for resilience and vulnerability of the communities as
communities and in terms of the place and importance of the commercial and recreational fishing
sectors within the communities. The ground-truthing research consisted of two major
components: (1) ethnographic field research within the communities, involving interviews with
fishers, with individuals whose businesses were related to fishing, and with community officials
and leaders about the place and importance of fishing within the communities; and (2)
compilation of historical and contextual background information to assess vulnerability,
resilience, and sustainability of the communities, especially through socioeconomic dependence
of the community on their fisheries and vulnerability from gentrification.
Each of the ground-truthing components produced data that allowed for relative ranking of the
nine communities on the Texas coast (the two South Atlantic communities were not included
here, but were the selected research sites for another related Foundation project). The interview
data were based on culturally significant categories provided by or derived from the interviewees
themselves. The interviews were coded for both fishery-level data and individual-level data,
allowing for community rankings on those two data types. Rankings were also constructed from
the historical/contextual information. The rankings from the two types of data were highly
similar. In general, the more diversified the economy of a community, the more resilient and the
less vulnerable the community. The most vulnerable communities were most reliant on one
economic factor, specifically commercial fishing.
The relative rankings for each of the eleven communities on the factor loadings were presented
in terms of vulnerability and resilience. The rankings from the factor loadings showed variation
9
across the communities, understandably, but systematically in relation to expectations of the
impacts of the factors toward vulnerability or resiliency. In addition, comparisons of the
combined ethnographic rankings with the quantitative were overall positive and statistically
significant. The ground-truthing thus confirmed that the indicators were, in fact, reliable
measures for the concepts under consideration.
IV. PURPOSE
A. Description of Problem
The FAO Technical Guidelines for Responsible Fisheries promotes the use of indicators to
monitor sustainability and other measures of well-being (Boyd and Charles, 2006). While there
has been substantial progress in the development and implementation of sustainable development
indicators for marine fisheries at the national and regional levels, “there has been little attention
paid to establishing frameworks at the local or community level” (Boyd and Charles, 2006:238).
Associated with the idea of sustainability are the notions of resilience and vulnerability, which
have seen increasing use with regard to coastal hazards at the community level (Cutter et al.,
2000), but not with regard to fishing communities. Yet, the recent devastation to Gulf Coast
fishing communities after hurricanes Katrina and Rita makes this form of vulnerability an
important aspect of the coastal ecology (Impact Assessment, 2006a). There has been increased
use of rapid assessment to profile coastal fishing communities (Jacob et al., 2002; Impact
Assessment, 2004; 2005a; 2005b; 2006; Jepson et al., 2005), yet there has been a modest
emphasis placed upon the ability to extract information for social impact assessment. Although
community profiles are important, they are only the initial step in a series of data needs to assess
the impacts of actions or cumulative impacts of regulation and other forces. This project, by
ground-truthing the initial construction of indicators and indices through more in-depth research,
reinforced their development and added to the social impact literature for fishery management.
In the NOAA Fisheries document “Guidance for Social Impact Assessment” (2001), there are
five categories of social factors identified for use in assessing management impacts. These
groupings are: 1) demographics for the community; 2) cultural data related to the fishery; 3)
social structure and institution impacts; 4) cultural data related to the community; and 5)
historical and current participation in the fishery. For each of these broad groupings, there is a
significant need to develop sources of secondary data to deal with budget and time constraints.
Some of these five groupings are more easily operationalized with secondary data using social
indicators. Social indicators are secondary data that are collected on a regular basis and
monitored to assess social conditions and well-being. This would include data such as
population size, number of residents living in poverty, or number of fishing permits. Social
indicators are often collected in absolute numbers and often longitudinal comparisons are made
within the community to assess change.
10
B. Project Objectives:
1. To convene a workshop to identify data and appropriate social indicators for construction
of social impact indices for fishing communities.
2. To develop a set of indices that will assist in the measurement of social impacts at the
community level for fishing communities of the Gulf and South Atlantic.
3. To place those indices into an information system to be linked to other indices of coastal
relevant information, i.e., measures of habitat sensitivity and/or coastal hazards.
4. To conduct an in-depth study of dependence, resilience, vulnerability, and sustainability
within selected communities, using large, secondary data sets and ground-truthing the results
ethnographically.
V. APPROACH
A. Description of the Work Performed
The results of a literature review, described below, provided the basis for selection of various
indicators included in the social impact indices. These measures were then discussed at the
planning workshop for inclusion in the models.
1. Vulnerability and Resiliency
There is an emerging consensus over the meanings of vulnerability and resiliency particularly
among human ecologists (Ahmed, 2006; Manyena, 2006). However, this is not to say that some
definitions in use are not contested (Ahmed, 2006:10). From a human ecological perspective,
vulnerability refers to the “differential susceptibility of ecosystems, households, or social groups
to losses” (Dow, 1999:76). A key to understanding vulnerability is that it can be applied on
different scales in the ecological model (Clay and Olson, 2008). For example, Cutter et al.
(2000) mapped the vulnerability of both people and places to natural and technological disasters.
They examined the differential vulnerability by social group, occupation, and location. In
fisheries, Smith and Jepson (1993) showed the vulnerability of Florida fishers, fishing families,
and consumers immediately prior to the state constitutional amendment that banned net fishing
in Florida in-shore waters.
According to Manyena (2006:436), a prevalent equation in disaster research involves the “risk =
hazard x vulnerability equation.” This risk equation is critiqued by Manyena (2006:436) as too
focused on what is out of the control of the community. Manyena (2006) suggests that the
concept of resilience helps focus actions on enhancing individual lives and communities rather
than the deficits people and places may have. This focus on assets and asset development is
opposed to a focus on needs or deficits. This would be a similar application as used in what is
called positive youth development with the concepts of assets and asset development creating
11
resiliency in youth to buffer life stress (Perkins and Butterfield, 1999). In addition, there is
corresponding literature in community development that relates to an asset development model
as opposed to a needs model (Allen, 2007).
The definition of resiliency is more contested than vulnerability, due in part to differing uses of
the term among disciplines (Ahmed, 2006). In ecology it is used to refer to how quickly an
ecosystem can return to a functional level after a disturbance and this is the definition we have
adopted (McEntire et al., 2002; Turner II et al., 2003; Clay and Olson, 2008). In other cases it is
used like the word resistance, as in resisting losses and damages when a disaster occurs
(McEntire et al., 2002:269). In other cases it is used in social-psychological theories to describe
how stressful life events such as divorce, job loss, and even disasters impact self-actualization
(Smith et al., 2003; Manyena, 2006).
When used in an ecological framework the concepts of resiliency and vulnerability are often
thought of as being linked on a continuum from vulnerable to resilient. Scholars have debated
the appropriateness of this coupling (Oliver-Smith, 1996; Ahmed, 2006). Theoretically,
vulnerability and resiliency should be different concepts because a vulnerable community does
not necessarily lack resilience and vice versa (Manyena, 2006). However, pragmatically they are
in fact often measured on a continuum with the assumption that vulnerable communities will be
less resilient and will need outside resources to recover from disasters (Buckle et al., 2001;
Manyena, 2006; Abel et al., 2007).
(A) The Relationship of Vulnerability and Resiliency to Well-Being
Cutter et al. (2000) make a distinction of social (really socio-economic) and ecological
vulnerability. For social vulnerability they examine population size, housing units, number of
females, number of minorities, people under 18 and over 65, mean house value, and number of
mobile homes (Cutter et al., 2000). These social characteristics are associated with a lack of
resources including information and knowledge, diminished power and representation, beliefs
and customs, and weak buildings, infrastructure, lifelines, and individuals (Cutter et al., 2000:
726). All of these factors when predominate in a community are thought to diminish well-being
as well as making the community vulnerable. In addition, such attributes are also thought to
diminish community resiliency as these factors decrease the amount of social interaction and
power these groups would have (Luloff and Swanson, 1995; Adger et al., 2005).
Ecological vulnerability is related to the risks associated with natural and technological hazards
(Cutter et al., 2000; Manyena, 2006). Hazards such as chemical releases, earthquakes, floods,
hail, tornadoes, hurricanes, severe wind, and wildfire are thought to increase ecological
vulnerability (Cutter et al., 2000). Most ecological systems can recover from natural disasters
and return to something resembling the pre-disaster state (Adger, 2000). However, certain
natural and technological disasters can completely change or destroy an ecosystem (Adger,
2000).
12
(B) The Time Element of Vulnerability and Resiliency
Both vulnerability and resiliency have an implicit time element. Manyena (2006) points out
indicators of vulnerability and resiliency are making a prediction about how a community would
be impacted if a natural or technological disaster occurred. Others have pointed out that fisheries
regulations also have widespread impacts on fishing-dependent communities. These regulations
should be considered along with natural and technological disasters and the vulnerability and
resiliency of communities (Jepson and Jacob, 2007; Clay and Olson, 2008). Regulatory change
fits in this framework because the regulations are a response to threats in the ecosystem (Clay
and Olson, 2008; Tuler et al., 2008). There have been no attempts to develop indicators of social
disruption that are contemporary with the disaster or regulatory change rather than predictive.
(C) Sustainability, Vulnerability, and Resilience
Associated with the idea of sustainability are the notions of resilience and vulnerability, which
have seen increasing use with regard to coastal hazards at the community level (Cutter et al.,
2000), but not with regard to fishing communities. Yet, the recent devastation to Gulf Coast
fishing communities after hurricanes Katrina and Rita makes this form of vulnerability an
important aspect of the coastal ecology (Impact Assessment, 2006a). The literature identifies
three primary forms of vulnerability/resiliency: 1) Social; 2) Economic; and 3)
Ecosystem/natural environment (Table 1). In addition, we have added social disruption. These
indicators are an attempt to capture more immediate aspects of social change rather than just
vulnerability and resiliency, which are predictive concepts.
Table 1: Factors that comprise community vulnerability and resiliency.
Social Vulnerability and Resiliency
Population Composition
Poverty
Housing Characteristics
Economic Vulnerability and Resiliency
Economic Structure
Ecosystem/Natural Environment Resiliency
Natural Disasters
Technological Disasters
Regulatory Impacts
Social Disruption
Housing Disruptions
Economic Disruptions
Personal Disruptions
13
(D) Measures of Social Vulnerability and Resiliency
Currently no social indicator index has been created with secondary data to establish social
vulnerability and resiliency. However, an index can be created through principal components
analysis to derive an index of social resiliency. Variables that should be examined for this index
include: family composition variables (including single-headed female households, percentage
of parents in the workforce with children under the age of six, retired households, households in
poverty, and households who primarily speak a language other than English), racial and ethnic
composition variables (percentage of racial and ethnic characteristics), income distribution and
poverty, age composition, and education levels (Table 2).
Table 2: Factors that comprise social vulnerability and resiliency.
Population Composition
Percentage Population White
Percentage Single-Headed Female Households
Percentage Parents In The Workforce With Children Under 6
Percentage Households That Speak A Language Other Than English In Home
Percentage Dependency Ratio (Under 18 And Over 65)
Percentage High School Degree
Percentage College Degree
Poverty
Percentage Households In Poverty 2007
Percentage Households 50% Under the Poverty Line 2007
Percentage Of People Over 65 In Poverty
Percentage Of Children under 18 Living In Poverty
Cost of Living Index
Housing Characteristics
Median Rent
Median Mortgage
Average Number of Rooms
Percentage Homes with Inadequate Plumbing
Average House Age
(E) Measures of Economic Vulnerability and Resiliency
Jepson and Jacob (2007) have developed an economic vulnerability and resiliency index for
fishing communities. One factor that was considered to be important while taking into account
impending regulation was the availability of employment within these communities.
Commercial fishers often engage in other types of work if fishing is slow or they face closed
14
seasons. In fact, most commercial fishers have employment histories that include extended work
outside of fishing, although they prefer fishing to most any other type of job. Previous research
had suggested that employment opportunities were not confined to the local community but often
encompass a more regional area (Jacob et al., 2002; St. Martin and Hall-Arber, 2008). The
measures in this index assess the local economic structure (Table 3).
Jacob and Jepson (2009) have developed a local community index based on the Fish Stock
Sustainability Index, which is calculated for the entire region but can be used at the local level to
indicate which communities are lacking diversity in their fishing income. Based on the
percentage of local landings as compared to regional landings a statistic identical to shift share
analysis can be calculated. This can also be used to examine the economic vulnerability and
resiliency of communities.
Table 3: Factors that comprise economic vulnerability and resiliency.
Economic Structure
Median Household Income
Unemployment Rate
Percentage Population In Labor Force
Percentage Self Employed
% Population Receiving Supplemental Support Income (SSI)
(F) Measures of Ecosystem / Natural Environment Vulnerability and Resiliency
The work of Cutter et al. (2000) uses both historical natural disaster data and the locations of
industrial activities that could potentially lead to technological disasters. Specific variables
included the number of recorded hurricane strikes, surges, and winds by category, the rate of
occurrence of 100 year and 500 year flood inundations, potential rail, highway, and fixed facility
accident zones, and earthquake occurrences (Table 4). These variables were compiled into a
single index and a score calculated for all communities in Georgetown County, South Carolina.
Later Cutter et al. (2003) developed what they termed the Social Vulnerability Index (SoVI).
This index furthered the disaster vulnerability research by including population factors such as
poverty, education levels, gender and age along with natural and technological hazards. This
was a significant improvement in that Cutter et al. (2003) showed that population factors could
mitigate or intensify the impacts of disasters. Cutter (2003 p.439) has focused on the use of
Geographic Information Systems (GIS) to display this differential vulnerability and to help plan
for the “emergency response cycle.”
It should be pointed out that this work lead by Susan
Cutter has been completed at the county level.
A quick way to assess vulnerability and resilience could also be addressed by habitat loss and
recovery (often through mitigation and offsets) seen in the NOAA Coastal Change dataset. This
would include wetland indicators and percent change in gross forms of land cover.
15
Table 4: Factors that comprise ecosystem/natural resource vulnerability and resiliency.
Natural Disasters and Technological Disasters
Damaging Hail Risk
Damaging Hurricanes Risk
Damaging Tornadoes Risk
Damaging Winds Risk
Environmental Protection Agency Registered Facilities
(G) Indicators of Social Disruption
Social Disruption is indicated by rapid change in population, income, and housing in
communities. This is related to natural disasters, boom and bust cycles, as well as gentrification
(Wilkinson, 1999). The key is capturing change over time and so these statistics are calculated
as percentage of change over a decade or other time period. Variables used in prior research by
Jacob et al. (2002) and Jacob et al. (2005a) include total population change, change in population
composition including age, race, ethnicity, and family composition. Housing, income, and
commuting patterns were also used including: change in the number of vacant homes, number of
rentals, length of commute, construction permits, median income over time, change in median
property value, unemployment rates, and changes in both commercial and charter boat permits
(Table 5) (Jacob and Willits, 1994; Jacob et al., 2002).
Table 5: Factors that comprise social disruption.
Housing Disruptions
Percentage Increase In Median Rent 1990-2000
Percentage Increase In Median Mortgage 1990-2000
Percentage Change In Median Property Value 2000-2007
Percentage Change In Renters 1990-2000
Percentage Moved Into Household 1999-2000
Percentage Moved Into Household 1995-1998
Personal Disruptions
Percentage Change in Unemployment
Percentage Change In Commuting Times to Work
Number of Registered Sex Offenders per 1,000 Population
Percentage Population Separated
Percentage Population Divorced
16
2. Natural Resource Dependency and Well-Being
Historically most communities begin as natural resource dependent and are founded so residents
can meet their sustenance needs (Hawley, 1986; Jacob et al., 2005a). As communities evolve
most develop diversified economies that are less dependent on resource extraction (Freudenburg
and Gramling, 1992). However, some communities do not diversify and it is thought that these
communities are more vulnerable to drastic economic fluctuation and instability (Stedman et al.,
2004). This is because: 1) resource values are determined by outside markets (they are price
takers); 2) there are external competitors who drive the price lower; 3) in most cases high prices
lead to product substitution; 4) technology increases efficiency, reducing the need for labor and
flooding the market; 5) the resource itself may be cyclical and abundance unknown; 6) resource
regulations; and 7) the resource can be depleted (Freudenberg, 1992; Smith and Jepson, 1993;
Peluso et al., 1994; Weeks and Packard, 1997; Stedman et al., 2004). It is also thought that
natural resource dependence makes communities less resilient to natural and technological
disasters (Donoghue and Sturtevant, 2007).
Not surprisingly a great deal of research has been conducted on the relationship of natural
resource dependency to individual and community well-being (Parkins et al., 2003; Stedman et
al., 2004). Researchers have found that the stress of working in extractive industries to be
detrimental to mental health when the resource is subject to any of the seven factors listed above.
For example, Smith et al. (2003) found that resource regulations and restrictions lead to
increased family stress and stress outcomes such as depression and anxiety for both fishers and
their partners. During the farm crisis of the late 1980s and early 1990s similar findings were
observed for farm families (Armstrong and Schulman, 1990; Belyea and Lobao, 1990; Jacob et
al., 1997). At the community level there has been a strong link between poverty and natural
resource dependency (Freudenberg, 1992; Peluso et al., 1994; Parkins et al., 2003; Stedman et
al., 2004).
(A) The Relationship of Resource Dependence on Vulnerability, Resilience, and Gentrification
Resource Dependence is thought to be associated with poverty and other measures of income
mal-distribution. The poor have fewer economic resources to handle the fluctuations in income
that typify resource dependency and as a result are very vulnerable (Peluso et al., 1994). At the
aggregate level communities find themselves vulnerable as tax revenues decrease while the
demand for services from the impoverished increases (Jacob et al., 2005b). Vulnerability can
also occur in the form of natural and technological disasters. However, one of the best indicators
of disaster preparedness relates to the mean income and educational level of residents (Cutter et
al., 2000). Communities that have diverse economies, higher levels of income, and education
tend to be less vulnerable and more resilient to economic fluctuations and seem to recover
quicker from natural disasters (Oliver-Smith, 1996; Flint and Luloff, 2005). Gentrification is
common in coastal areas that are fishery dependent. Many of the cultural factors of fishing are
attractive to outsiders and so fishing communities often experience increases in property values
17
that squeeze locals – both fishers and those in related industries – through higher taxes (Jacob et
al., 2005a). In addition, gentrification may make communities more vulnerable and less resilient
(Donoghue and Sturtevant, 2007).
(B) Fishing Community Dependency and Social Impact Assessment
Applied anthropologists have had a long-standing interest in commercial fishers and fishing
communities (Acheson, 1981). This interest has led to a drive to understand exactly how
commercial fishers and their communities endure changing natural and social environments,
including a growing interest in regulatory impacts (Acheson, 1981). Regulatory impacts have an
obvious impact on fishers and communities (Smith and Jepson, 1993). These impacts have been
most easily seen in dramatic examples of fishery closures (Smith et al., 2003; Clay and Olson,
2008; Tuler et al., 2008). However under the revised Magnuson-Stevens Fishery Conservation
and Management Act National Standard 8, federal policy now mandates that fishery
management plans identify and consider the less dramatic social and economic consequences of
fisheries management actions on fishing communities, to assure their sustained participation and
minimize adverse impacts (MSFCMA Section 301 [a][8]). This mandate is based on the
recognition that conservation and management efforts have expansive social and economic
impacts in fishing communities, affecting not just the individual harvester or processor, but also
impacting “directly related fisheries-dependent services and industries" (for example, boatyards,
ice suppliers, tackle shops)(Federal Register, 1998). The Act defines fishing-dependent
communities as “a community which is substantially dependent on or substantially engaged in
the harvest or processing of fishery resources to meet social and economic needs, and includes
fishing vessel owners, operators, and crew and United States fish processors that are based in
such a community ” ((Magnuson-Stevens Act, section 3(16)).
Impact Assessment Inc. (2004, 2005a, 2005b, 2006a, 2006b) in a series of reports written to
profile fishing communities on the Gulf of Mexico developed a three category typology of
fishing dependence. The categories were 1) Primarily-Involved, 2) Secondarily-Involved, and 3)
Tangentially-Involved. The communities classified as “Primarily-Involved” are seen as the most
vulnerable to changes in fisheries regulations, economic fluctuations, or other perturbations in
the fishery because this is the primary economic and social focus of the community. Those
classified as “Secondarily-Involved” still may have substantial social and economic impacts but
the community and economy are more diversified and less vulnerable. Those communities
“Tangentially Involved” are the least vulnerable.
Federal law now mandates social impact assessment of fisheries regulations including
allocations, reallocations, closures, restrictions, limited entry schemes, or any other policy
change that might adversely impact fishing-dependent communities (Jacob et al., 2001; Clay and
Olson, 2008). Here we hope to clarify the concept of dependence so appropriate communitylevel indicators can be developed to assist in impact assessment. It is important that these
18
indicators cover the range of the concept. Some critical components of dependence are included
in Table 6 below.
Table 6: Factors that comprise forms of fishing dependence in communities.
Economic Dependence
Commercial fishing dependence
Recreational fishing dependence
Non-consumptive value dependence
Social Dependence
Institutional dependence
Cultural heritage dependence
Landscape and built environment dependence
Socioeconomic Dependence
A cumulative ranking of all of the above
(C) Dependence and Dominance in the Local Community
Jacob et al. (2005) explored the linkage of differing forms of dependence and their relationship
to dominance. In local communities it is relatively easy to understand the link between
economic dependence and the local power structure. However, it is a bit more difficult to
understand how institutions that evolve from resource extraction become dominant in the local
community. Important cultural institutions such as the educational system can be dominated by
resource extraction concerns. For example students may be trained to work in extractive
industries rather than developing other forms of human capital. Other institutions can be
dominated as well, such as political, religious, economic, and kinship systems which all can
reinforce the dominance and dependence of the extractive activities of the resource.
(D) Dependence versus Reliance
Natural resource dependence has usually been defined by economic data with a specific cutoff
point, such as 15% of total jobs or income from fishing defining dependence (Frere and Failler,
2001; Jacob et al., 2001). Additionally, economic multipliers are often used with natural
resource income to capture the forward and backward linkages in the local economy that relate to
fishing (Frere and Failler, 2001; Jacob et al. 2001). Jacob et al. (2002) also suggested a
community is fishing dependent when some substantial figure from fishing (ranging from $10
million to $100 million depending on population size) is reached, even if the total percentage of
income from fishing is below 15%. This purely economic approach creates a dichotomy of
fishing-dependent communities and communities that do not meet the a priori threshold
19
(Stedman et al., 2004). There are some substantial criticisms of this approach. First, this
dichotomy is insensitive in that it may not substantively discriminate a community that is just
under the threshold to one that is just over. Further non-consumptive uses of the resource have
been left out. In addition, it is disappointing that social dependence has been completely missed
(Clay and Olson, 2008). This is no doubt due to the fact that it is very difficult to quantify social
fishing dependence. We propose a variety of measures of economic and social dependence that
will preserve the range of outcomes to maximize variation among the communities. Each
indicator will be ranked and these rankings will be summed or factor analyzed to complete an
index of fishery reliance for each community (we will conduct analysis to determine the best
approach). The concept of reliance will give us the maximum variation without losing
sensitivity or “throwing data away” through the coding process (Stedman et al., 2004). This in
turn will allow us to quantitatively explore the relationship of fishery dependence with
vulnerability, resilience, and gentrification.
(E) Potential Indicators of Fishing Dependence
(1) Economic Dependence
Commercial and recreational dependence is generally defined as low as 15% or greater of jobs or
income coming from commercial and recreational fishing and related industries. Often
economic multipliers are used as well to capture the backward and forward linkages in the
economy. In addition, our research will attempt to capture non-consumptive values. In our
database we will have the following variables available to test, develop, and construct measures
of economic dependence.
Table 7: Potential measures of commercial, recreational, and non-consumptive economic
dependence.
Commercial Dependence Indicators
Percentage Labor Forces Employed in Agriculture, Fishing, and Hunting
Pounds of Landings per 1,000 persons
Commercial Fishing Permits per 1,000 Population
Value of Landings per 1,000 Population
Dealers With Landings per 1,000 Population
Recreational Dependence Indicators
Charter Boat Permits per 1,000 Population
Marinas and Related Businesses per 1,000 Population
Marinas and Related Businesses Jobs per 1,000 Population
Marinas and Related Businesses Gross Income per 1,000 Population
Boat Launches per 1,000 Population
20
(2) Social Dependence
There is a dearth of indicators that measure social dependence on fishing within a community.
In this research we will rely on indirect indicators to measure the concept. Social indicators can
be direct or indirect. A direct indicator is one where the measure is the variable of interest. For
example, if one were interested in the level of health of children in a community, a medical
examination of a sample of those children in the community would be a direct measure (Rossi
and Gilmartin, 1980). Direct indicators specifically measure the immediacy of the interest.
Indirect indicators measure a variable correlated to the variable of interest, not the concern itself.
Indirect indicators are measures based on experience or theorized to be related to a variable of
interest. Again, if the variable of interest were the general health of children in a community, an
indirect indicator would be the school absentee rate (Rossi and Gilmartin, 1980). Direct
indicators are generally preferable to indirect indicators; however, direct indicators are not
always available.
Table 8: Potential indicators of social dependence.
Social Dependence Indicators
Percentage Water Cover in the Municipality
Boat Launches per 1,000 Population
Percentage Labor Forces Employed in Agriculture, Fishing, and Hunting
Marinas and Related Businesses per 1,000 Population
Dealers With Landings per 1,000 Population
3. Gentrification and Well-Being
Gentrification is classically defined as the displacement of lower or working class residents by
the middle and upper classes (Hamnett, 1991; Atkinson, 2000; Lees, 2000; Wildin and Minnery,
2005). Early work in the study of gentrification focused on urban neighborhoods or districts
near the central business district that had become rundown (Atkinson, 2000). Because of the
high costs of housing and commuting, the rundown neighborhood becomes attractive (Hamnett,
1991). The new residents are thought to move in and rehabilitate existing structures to meet their
middle and upper class standards (Lees, 2000). The initial wave of newcomers typically has
more middle class aspirations than financial capital and so they tend to invest a great deal of
“sweat equity” into their homes (Lees, 2000). Successive stages of newcomers are wealthier and
pay contractors to have the houses upgraded and sometimes even displace the early gentrifiers
(Lees, 2000). As improvements are made throughout the neighborhood or district the area
becomes more attractive to investors and outsiders and property values increase (Atkinson,
2000). Eventually services and consequently property taxes increase as a result of the
improvements and this often pushes out the lower and working class residents (Atkinson, 2000).
21
Gentrification is a loaded term because of the class distinction and the tenure insecurity the poor
have even when they own property (Atkinson, 2000). Additionally class distinctions are often
entangled with age, gender, racial and ethnic status, which makes the process of gentrification
even more controversial as it seems very exploitive (Lees, 2000). The very character of
gentrified neighborhoods are thought to change as the new residents attract amenities such as art
galleries, bakeries, bistros, coffee shops, and martini bars which displace bodegas, sub shops,
and bars (Zukin, 1995). In addition, the occupational structure of residents shifts from a bluecollar base to services and white-collar professionals (Atkinson, 2000). This cultural shift
completes the total change of both space and sense of place. Those members of the lower and
working classes who remain are no longer able to relate to their community as they once did.
They also struggle with higher rents or taxes and cost of living. But what happens to those that
are dislocated by this process?
Those who are displaced generally do not have enough resources to purchase new homes in tight
urban housing markets, even with the increased value from gentrification (Atkinson, 2000).
Those who are renting also face similar difficulties and often relocate to even less desirable
locations (Atkinson, 2000). The same market factors that make neighborhoods attractive for
gentrification, such as scarce and expensive housing, make life very difficult for the displaced
(Atkinson, 2000). So how does gentrification impact community well-being? It undoubtedly
depends on your point of view. Administrators, developers, contractors, and the middle and
upper classes view gentrification as a cure for almost every city problem while those who are
displaced are far less enthusiastic (Atkinson, 2000). Awareness of social justice issues related to
gentrification has highlighted the fundamental need for affordable housing that would improve
overall community well-being (Lees, 2000). There is less awareness about preserving the
existing character of neighborhoods and now there is actually a movement by cities to “theme”
districts to reflect the historical character of a place, but this “theming” of place is ultimately
very inauthentic (Chang, 2000).
(A) Etiology of Gentrification
There are competing theories about the source of gentrification pressure in cities. Lees (2000)
discusses the city as an emancipatory space that is attractive to the “new middle class.” The new
middle class are typically young, educated, and childless and seek the liberating lifestyle that
urban places offer (Lees, 2000). In some studies, gentrification by gays, lesbians, women, and
other minorities reaffirms the emancipatory pull of the city (Lees, 2000). Since they are “new
middle class” they may not be able to afford established middle or upper class neighborhoods in
the city and thus tend to pioneer gentrification efforts (Lees, 2000). This argument focuses on
the agency of the individuals involved in gentrification and less so on the real estate market
forces (Lees, 2000).
Others have focused more on the supply side economics of gentrification (Hamnett, 1991; Lees,
2000; Wildin and Minnery, 2005). Simply put, a housing shortage motivates young
22
professionals to move into city neighborhoods where they find good housing stock and
neighborhood “potential” relatively inexpensive (Wildin and Minnery, 2005). This is in turn
followed by a second stage of more established middle-class who further bid up property values
as the area becomes more desirable (Lees, 2000). This process is cyclical and may even involve
re-gentrification over a period of time (Lees, 2000).
Still others identify the role of culture and consumption in post-industrial society as a driving
force in gentrification. With the increasing loss of manufacturing in cities, former spaces of
production are converted into bars, boutiques, galleries, and even lofts and condominiums
(Zukin, 1995). The cultural heritage of manufacturing is transformed into places of consumption
of high culture that has displaced the working class social structure (Lees, 2000). This has an
agglomerating effect as the gentrification process takes root and grows, attracting new services
and businesses, and further displacing the older existing businesses. This in turn attracts
wealthier residents and displaces long-term poorer residents.
(B) Urban versus Rural Gentrification
Lees points out that gentrification is not the same everywhere (2000). In cities, suburbs, and
rural places one could expect varying processes to drive gentrification to differing outcomes. In
the urban case gentrification operates under a fairly straightforward set of dynamics including
market forces, individual agency, and cultural preferences towards consumption. These forces
are readily observed in the compact space of a city neighborhood or district. Unfortunately,
much less is known about exurban or rural gentrification; “research into its causes and
consequences has been lacking” (Yagley et al., 2005:1). Rural gentrification is thought to
encompass the suburbanization and urban sprawl phenomenon that has accompanied the
depopulation of cities from the mid 20th century to now. The phenomenon is so coercive and
widespread that we often fail to recognize the impact on local society surrounding cities (Yagley
et al., 2005). But exurban gentrification exists too as people are attracted to small town and rural
life and leapfrog the suburbs entirely (Bell, 1992; Phillips, 2002). Importantly, the major
difference between urban gentrification and rural gentrification is that urban gentrification deals
largely with the rehabilitation of existing structures while rural gentrification often (but not
exclusively) includes a great deal of new development (Phillips, 2002; Yagley et al., 2005).
(C) Forms of Rural and Coastal Gentrification
Yagley et al. (2005, 1) identified three scenarios where rural gentrification was likely to occur: 1)
urban sprawl; 2) people attracted by natural amenities; and 3) retirees attracted by low cost of
living and environment. These factors are obviously very different from those that impact urban
gentrification. Still, market forces are in play as cost of living and real estate tend to be cheaper
the further one gets away from the central city and this certainly attracts new residents.
Additionally personal preference and agency also play a large role, as some simply prefer large
lots, rural landscapes and access to nature.
23
One of the most problematic outcomes of rural gentrification includes increasing housing costs.
This troubles long-term rural residents because it makes it difficult for their children and
extended family members to live nearby, as had been the norm in the past. The young in the area
get priced out of the home market and have to relocate further away from family. This
undermines the traditional family values that have been documented in rural community life
(Jacob et al., 1997; Wilkinson, 1999; Jacob et al., 2005b). In addition increasing property values
increase local taxes, which may eventually displace long-term residents, especially the elderly
(Phillips, 2002; Yagley et al., 2005). Other long held traditions also change in the “. . . local
culture in exchange for a homogenous, suburbanized, new identity” (Yagley et al., 2005:1).
Further, the age, economic, and political structures also undergo significant change as wealthier
residents move into town (Phillips, 2002; Yagley et al., 2005). Community economic activity
often shifts from natural resource production to services (Yagley et al., 2005). Newcomers often
push for the development of governmental services and regularly become politically involved as
they often have the time, resources, and education to achieve their ends (Yagley et al., 2005).
Newcomers are often welcomed in communities by elements of the “local growth machine” such
as realtors, developers, utilities, and newspapers (Molotch, 1976). An argument used by these
local elements for rural gentrification is that the new residents will often generate demand for
service jobs and research does confirm the increase in service employment (Phillips, 2002;
Yagley et al., 2005). However service jobs often pay poorly, offer little advancement, provide
few benefits, and are more vulnerable to economic downturns than other job sectors (Yagley et
al., 2005). In many cases long-term residents are less excited about the newcomers (Jacob et al.,
2008).
Coastal gentrification, like rural gentrification has received relatively little attention. Like rural
gentrification, coastal communities would be susceptible to the same three factors that drive rural
gentrification: 1) urban sprawl; 2) people attracted the natural amenities; and 3) in-migration of
retirees (Table 12) (Yagley et al., 2005). In many cases coastal communities may face all three
factors simultaneously. Jepson (2004) in his study of Cortez, Florida documented the occurrence
of all three factors in addition to a large influx of tourists. The outcomes of coastal gentrification
are also similar to rural gentrification with the addition of concerns about access to the
waterfront, especially for commercial fisherman (Maine Sea Grant, 2007; Hartley et al., 2008).
Maine Sea Grant (2007:3-7) conducted research on the consequences of uncontrolled coastal
development and detailed the following consequences: 1) loss of access for commercial fishers;
2) recreational fishing access conflicts with commercial fishers and other users; 3) limited public
access; and 4) environmental impacts on important ecosystems. Khan (2007) detailed the
multiple threats to ecosystems from uncontrolled coastal development.
24
Table 9: Factors that comprise forms of coastal gentrification in communities.
Urban Sprawl
People Attracted To Natural Amenities
In-Migration Of Retirees
(D) In-authenticity of Coastal Gentrification
When urban gentrification takes place many older building facades, businesses, and traditions are
changed in favor of those preferred by middle and upper class residents. There is little attempt to
keep the material culture of the long-term residents (Lees, 2000). During coastal gentrification,
there is some attempt to keep a fishing heritage “theme.” Jacob et al. (2005a) detail this fishing
heritage “theme” in the built environment and also discuss the role of community narrative
rhetoric in reinforcing this “theme.” However, as Chang (2000) points out that such “themed”
space tend to be inauthentic. In fact Chang (2000) described how a themed landscape tamed
Singapore’s “Little India.” From Chang’s work three “taming” generalizations can be made that
could apply to coastal gentrification: 1) traditional activities decline in favor of staged activities;
2) the conversion of cultural significant production activities are shifted into consumption
amenities; and 3) a rich cultural heritage becomes a caricature. This taming has the effect of
making commercial fishing activities very difficult in the community (Jacob et al., 2005a).
(E) Potential Indicators of Coastal Gentrification
(1) Urban Sprawl Indicators
Urban sprawl is dependent upon proximity of a city that serves as a central place for employment
and a population reservoir that fuels sprawl as the central core decentralizes into multiple cores
outside of the city center (Gottdiener and Budd, 2005). When sprawl occurs population
increases while population density actually decreases (Table 10). This is because new housing is
likely to be developed on large parcels. As the sprawl increases, median income usually
increases. The number of new residents that lived outside of the county five years ago should
capture the number of residents who are attracted to the multiple cores that now surround the
city. Commuting times are likely to increase but commuting locations should be into the center
city or one of the multiple cores surrounding the city. Median home values should increase
along with an increase in affordable housing that helps offset the gentrification.
25
Table 10: Indicators of urban sprawl.
Nearest City with 50,000 or greater population
Changes in population density 2000-2007
Percentage of homes built between 1995-1998
Percentage of homes built between 1999-2000
Percentage new residents in the last 5 years (lived in a different county 5 years ago)
Percentage of homes less than $100,000
Cost of living index
(2) Natural Resources Migration
When new residents are attracted to a community with high levels of natural resource amenities
(such as views, water access, and productive sports fishing) there is often a percentage decrease
in natural resource production and a concomitant increase in services. Many natural resource
dependent communities have a higher number of mortgages that are funded by Farmers Home
Administration (FHA) or USDA Rural Development funds. High amenity communities would
also have a great deal of natural land cover (water, forest, farm) as well as land that is preserved
in parks or reserves. The number of businesses in resource extraction can also indicate the
existence of resource amenities. Absentee owners indicate the number of second homes or
vacation homes that exist because of the amenities. A high number of rentals in the community
also reinforces this idea.
Table 11: Indicators of natural amenities migration (people attracted to natural amenities).
Percentage of homes rented
Percentage of homes vacant
Number of boat ramps within the municipality
Percentage of water cover within the municipal boundaries
Percentage of labor force in agriculture, farming, fishing, and mining
(F) In-Migration of Retirees
When retirees migrate to a community, population naturally increases, but more importantly, the
age structure within the community changes greatly. Additionally the governmental transfer
payments from social security produce an outside income source in the community. To
indirectly measure the increase in retirees we can look at the number of nursing home beds per
10,000 population.
26
Table 12: Indicators of in-migration of retirees.
Percentage population over age 65
Percentage population receiving social security
Mean retirement income
Percentage labor force in services
Number of nursing home beds per percentage of population
4. Triangulation: Ground-Truthing
The research design of the project called for ground-truthing of the characterization of the
selected communities from the large, secondary data sets. The ground-truthing part of the
project was designed to independently test for resilience and vulnerability of the communities as
communities and in terms of the place and importance of the commercial and recreational fishing
sectors within the communities.
The ground-truthing research consisted of two major components:
•
Ethnographic field research within the communities, involving interviews with fishers,
with individuals whose businesses were related to fishing, and with community officials
and leaders about the place and importance of fishing within the communities.
•
Compilation of historical and contextual background information to assess vulnerability,
resilience, and sustainability of the communities, especially through socioeconomic
dependence of the community on their fisheries and vulnerability from gentrification.
B. Project Management
1. Planning Meetings
(A) Social Indicators Workshop
The workshop was held at the Houston Advanced Research Center on June 15-17, 2008. The
invited participants for this workshop were: Dr. Blount, Dr. Weeks, Dr. Jacob, Dr. Jepson, Dr.
Alyne Delaney, Dr. Richard Pollnac, Dr. David Griffith, Dr. Manuel Valdes-Pizzini and
Technical Monitor Dr. Palma Ingles, who are all experts in the study of fishing communities.
On Sunday, June 15, 2008, a field site visit was conducted. Those who participated in the field
site visit were: Dr. Delaney, Dr. Jacob, Dr. Valdez-Pizzini, Dr. Pollnac, Dr. Blount, Dr. Ingles,
Dr. Weeks and Dr. Jepson. The group drove to several communities on Galveston Bay to better
understand how the fishing infrastructure was incorporated into the larger community. It became
clear that, along Galveston Bay at least, there were no discrete fishing communities, but enclaves
that were surrounded by industrial, residential and/or recreational tourism development. The
27
group discussed this scenario and how the interns would approach their interviewing and
documenting of the fishing infrastructure.
The workshop formally began on Monday, June 16. Dr. Jepson gave a brief introduction to the
project explaining how a community vulnerability index was constructed for the Gulf Council’s
EFH amendment EIS. It was pointed out that the goal of the research was to develop several
indices that would then be able to be used in social impact assessments for future fishery
management.
Dr. Alyne Delaney, Aalborg University Research Centre, Denmark, discussed her work with the
North Sea cod and hake recovery plans. There were 3 communities in which she had been
working, two of those communities are: Peterhead, Scotland – rely on oil and gas but don’t go
back and forth; Urk, Netherlands – Calvinist and extremely conservative. The types of data they
were collecting are fisheries sector - landings, support sectors, some of the demographics – age,
gender and ethnicity seem to be important variables that are tied to social impacts. Some of their
focus was trying to understand community support and unemployment, more specifically looking
at social networks and social capital (fewer boats so fewer social networks). The management is
through the Common Fisheries Policy – each member state has a percentage of the quota.
Heritage tourism is becoming a focus for many communities. She noted that some of the data
collected was coming from web pages for businesses and other entities associated with fishing.
Dr. Richard Pollnac, University of Rhode Island, worked with northeast fishing community
profiles and had assembled over 100 fisheries variables. Using those variables, he assembled a
typology of fishing communities through a principal component analysis using 43 fisheries
variables. He has conducted job satisfaction work around the world as a measure of well being –
the name had to be modified to activity satisfaction to accommodate the recreational component.
Activity satisfaction attributes are connected to both individual participation and social conflicts.
He did a principal component analysis of satisfaction variables and found that self-actualization
is an important variable. He had also considered the idea that fishermen may be genetically
disposed for the occupation. They are drawn to risky behavior, like fishing. When you look at
satisfaction, one of the highest loading variables was mental stress.
Dr. David Griffith, East Carolina University, began his presentation by pointing out that most
research on fishing communities to date had been ethnographic overview and often rapid
assessment. In describing his work in Puerto Rico, he described how they used an 8 item index
of dependence: community type, ratio of full-time/part-time fishers; ties to tourism; involvement
in coastal conflict; ties to the state; fishing infrastructure; ceremonial infrastructure/activity; rank
in landings (scale of 1-5). He went on to suggest that the social indicators approach can serve as
a basis for community health over time; can be figured before and after; and makes sense to
managers. Some of the drawbacks are that it takes items out of context; it is static (communities
change); conceptualizes communities as systems or closed units of analysis; feeds the “crisis of
representation” critique; and it is difficult to quantify the meaningful sense of some important
28
attributes of fishing. He suggested a complementary approach that uses cultural biographies;
conceptualizing households and communities as ongoing processes rather than systems, with a
focus on a local setting.
Dr. Manuel Valdes-Pizzini, Interdisciplinary Center for Coastal Studies, Recinto Universitario de
Mayagüez, began by suggesting that fishing communities of St. Croix are not place based, but
are network based. Much of the fish coming into St. Croix is through imports and not the local
fishing. Most fishermen are part-time. Historically, cod fish and herring were important; slaves
were important in the fishing trade and women’s involvement in marketing was common place.
It all changed in the twentieth century when the Puerto Ricans came to the island. His point was
that we cannot understand these fishing communities without understanding their history. He
suggested an analysis of the historical background of the communities that are chosen should be
conducted that as part of this research.
Dr. Steve Jacob, York College, began by asking the group as to whether the communities should
be chosen through a data driven approach, typological constructs or post hoc rationale. Dr.
Weeks pointed out that the communities in Texas were chosen through a more typological
approach in that the Galveston area is more urban and gentrified, whereas the communities in the
San Antonio Bay area are more rural and less gentrified. Dr. Jacob decided that pairs of
communities would be chosen using the more typological approach. He then brought up the
appropriate level of analysis and what should that unit be: MCD, census block, CDP...etc. It was
suggested that CDP be used since it had been used previously in other studies in the Gulf and
South Atlantic. However, it was noted that in some cases, communities were combined to allow
for a better overview of the economic, social and cultural extent of the fishing community.
On June 17, Dr. Jacob continued the discussion on the topic of dependence. The group discussed
several different variables that might be used for dependence, citing the work done by Impact
Assessment. The group discussed the difference between reliance vs. dependence. Dependence
implies it is a majority contributor to the economy, jobs, culture, etc., while reliance implies an
important contributor but not necessarily a majority. The importance of a historical perspective
was stressed and where possible data over several time periods is preferable. Measures of
vulnerability were discussed next. Employment opportunities are one variable that Dr. Jacob
used in the index. Lack of diversity in the economy was thought to make communities
vulnerable. The Fish Stock Sustainability Index was used at the local level to indicate which
communities are lacking diversity in their fishing income. Professionalization in recreational
sectors and over-investment in capital can be vulnerability indicators. It was decided to develop
a regulations vulnerability index and treat it as a Guttman Scale.
Measures of social change would include commutation patterns, the dependency ratio, poverty
levels, and other indicators that capture a vulnerable population, particularly over time. Other
measures of vulnerability would include community risk factors, economic disadvantages,
educational disadvantages, risks to disasters (flood maps, hurricane maps, and other FEMA
29
maps), vulnerable technologies such as refineries, nuclear power, and chemical plants, and
measures of social disengagement such as suicides, domestic violence, and DUIs.
There was a discussion about resiliency being just one end of a continuum with vulnerability at
the other, however, it was noted that there are measures of resiliency that are different enough
from being vulnerable that it should constitute its own index. Measures of social networks, like
number of different fishing associations and other social groups might be used as part of the
resiliency measure. Resiliency is relevant at the ecosystem, community, and family levels.
Changing regulations that reflect increasing health in species is an indicator of ecosystem
resiliency. Fishers remaining in fishing is a sign of resiliency.
Dr. Blount gave a presentation on cultural models as the basis for conducting the groundtruthing
exercise. He pointed out that the research is mostly qualitative and is sometimes difficult to
translate into a quantitative measure that makes sense to others. He reviewed issues of validity
with interviewing, and demonstrated ways to ensure increased validity across interviews and
interviewers. He outlined the procedures by which one will elicit certain key words from
interviews that will then begin to form cultural domains. It is the job of the interviewer to elicit
these key words and understand their meaning. He gave an example from work he did in
Georgia with African-American shrimpers and demonstrated how the model would be structured
based upon the key words.
(B) Social Indicators Workgroup Meeting
A workgroup meeting comprised of Dr. Steve Jacob, Dr. Priscilla Weeks, Dr. Ben Blount and
Dr. Michael Jepson was held at Flagler College, St. Augustine, FL, October 23 and 24, 2008.
The meeting began with Dr. Jacob reviewing his work on community boundaries. The group
discussed whether to use zip code data or census block data in the analyses. Because the
community of Port O’Connor, TX was not a census designated area, it was decided that zip code
boundaries would be used for those communities that were not census designated communities.
In the discussion about community boundaries, Dr. Weeks brought up a point about how land
gets fragmented through zoning and how that might be a key indicator of impending
gentrification. This was certainly apparent in Port O’Connor with the surrounding land
development. This point led to more discussion about the transition from public to private land,
especially with regard to the loss of public docks. If measureable, the availability or lack of a
public dock could be a sign of vulnerability.
Dr. Jacob then reviewed his documents on Fishing Dependence, Social Dependence, Resilience
and Vulnerability. He said there was considerable literature on the topics and mentioned he had
some questions on non-consumptive dependence. After some discussion, it was decided that he
would continue to look for such measures, although some of the suggestions for ecosystem
services value would be hard to operationalize. He continued with a discussion about
transportation as an indicator of gentrification and that he would explore some means of
30
developing a measure of such. He felt that he could easily find a measure of proximity to
interstates that would be easy to gather.
Dr. Blount reviewed his analysis of the qualitative data and discussed how they arrived at the
analysis. It was agreed by the group that his approach would be beneficial and was asked to
continue to refine the analysis by possibly collapsing some categories and split the Galveston
communities into discrete communities.
Finally, an outline of the final report was decided upon.
(C) Final Planning Session
The meeting was held on April 22 and 23, 2009 in Tampa, FL and was attended by Ms. Judy
Jamison, Mr. Frank Helies, Ms. Gwen Hughes, Dr. Blount, Dr. Weeks, Dr. Jacob, Dr. Jepson,
Dr. Ingles, and Mr. Bob Sadler. The first order of business was a short description of the project
and update on the progress to date. Some administrative issues were covered, primarily Mr.
Sadler discussing a no-cost extension for the project, which was agreed upon and later approved
by the NOAA Grants Office.
Dr. Weeks discussed the Galveston Bay communities studied by herself and the interns. It
consisted of interviews, photos of infrastructure and GIS mapping of the study areas. Dr. Blount
discussed the San Antonio Bay study area. He created a tabulated chart based on coded
interview data. He mentioned it needed to be cleaned up some and the ultimate goal was to
compare the qualitative and quantitative data to determine congruency. Dr. Jacob felt satisfied
with the correlative use of the differing year’s data from his quantitative analysis. The group had
some issues with defining vulnerability and resiliency and how fishing dependence relates to the
two metrics.
The Final Report draft was partially completed, but was still missing some analysis and
discussion. The group decided to collaboratively finish the paper with Dr. Blount being the
ultimate editor.
The group discussed lessons learned during the project. There was a discussion about Dr.
Jacobs’s ability to get clearance for additional data and the time frame involved in that. There
was a long discussion about future research. The group wanted to gather data on the household
level to better census the fishing communities and focus more on the effects on the families of
fishers.
2. Methods and Data
(A) Social Indicators
(1) Social Indicator Indices Development Strategy
31
The research leading to development of social indicators was conducted by Dr. Jacob. The
research was also discussed by the research group and feedback provided by them.
Three steps were taken to develop the social indicator indices. First, correlation coefficients
were examined to find underlying patterns of variation. Second, the variables that were most
highly intercorrelated and reflected the range of ideas of interest were placed in a principal
components analysis, where these variables were determined to be reliable indices. Last, the
variables were standardized and weighted for their effects in the model. Index factor scores were
used. Factor scores are similar to composite scores, with the exception that the items are
standardized and weighted in regard to their factor loadings. The factor loadings are a rough
indication of correlation of the domain concept’s latent structure to the single variable.
Therefore items that are most important in an index receive a higher weighting than a less
important item. In principal components, factor loadings less that .350 are generally not
considered to be significant and in most cases should be removed from a factor scale. One
advantage of factor scaling is that negative relationships do not have to be reverse coded before
scaling. This means that negative factor loadings work to reduce the overall score and the
absolute number conveys the strength of relationship regardless of being negative or positive.
The interpretation of a negative factor loading is similar to a negative Pearson’s r bivariate
correlation. The factor scores were standardized with a mean of zero and the scores reflecting
standard deviations from that mean. Scales were subsequently tested for internal consistency by
using Armor's (1974) theta reliability for factor scales. The theta coefficient is interpreted
similarly to Cronbach's Alpha, and is used for factor scales because it does not assume that all
items are weighted equally in the scale. Theta is calculated as: θ = [p/(p-1)]*[1-(1/λ)], where p =
the number of items in the scale and where λ denotes the largest eigenvalue from the principal
component analysis.
(2) Social Indicator Indices Components and Internal Reliability
To establish internal reliability, multiple indicators for each concept are necessary. At a
minimum it is necessary to include enough variables to fully cover the range of the concept,
while maintaining unidimensionality (only measuring one central concept). In general, multiple
measures are preferred and do increase internal validity when the items are significantly
intercorrelated. However as more variables are added to the index it is harder to maintain
unidimensionality. Unidimensionality, in part is established by principal components analysis.
In a principal components analysis a single factor solution provides evidence that the various
index items only measure a single concept. The indices in this study range from a low of four
items to a high of seven items. Indices with three or fewer items are generally thought to be
insufficient to establish internal validity through Cronbach’s Alpha or Armor’s Theta. Below
you will find a description of the components of each index, the principal components analysis
and factor loadings, and measures of internal validity including the eigenvalue, percentage
explained variation, and Armor’s Theta Reliability.
32
(3) Community Selection and Inclusion in the Data Set
There were three groupings of communities primarily based on estuaries that were selected for
study. In the Gulf of Mexico, on or near Galveston Bay, were the communities of Seabrook, San
Leon, Galveston, Texas City, and Bacliff. In the Gulf of Mexico, on or near San Antonio Bay,
were the communities of Port Lavaca, Sea Drift, Port O’Connor, and Palacios on Matagorda
Bay. In the South Atlantic were the communities of Little River and Mount Pleasant. These 9
communities were located in or adjacent to the following counties in Texas: Brazoria, Calhoun,
Chambers, Galveston, Harris, Jackson, Matagorda, and Victoria.
In South Carolina the
communities were located in or adjacent to Charleston, Georgetown, and Horry counties.
Because 11 communities would not provide sufficient variation in the data for reliable index
development it was decided to include all communities in the county and adjacent counties. This
resulted in a data set with 122 different communities and provided sufficient variation for index
development.
(4) Data Set Characteristics and Sources
The data set in this research was compiled from six separate data sources. The primary source
for population and housing information was the U.S. Bureau of the Census 1990 and 2000
Decennial Census, Summary Tape File Three. The main source for information about fishery
landings, permits, and value was provided as a custom database by NOAA Fisheries personnel.
Very current population estimates and real estate value was downloaded from the website “CityData.com.” The data for natural disaster risks was downloaded from “Moving.com.” The data
for technological disasters was downloaded from the Environmental Protection Agency’s (EPA)
website from the “EnviroMapper” store front. Last, data for marinas and related business were
downloaded from the 2002 Economic Census on the U.S. Census Bureau’s “American
FactFinder” web page. The data source and variable manipulation will be detailed below in each
index description.
(a) Issues with Confidentiality
An important issue in using community-level landings data revolves around federal
confidentiality rules. NMFS does not allow reporting landings data when there are less than
three fishers, processors, or distributors in a given community (Impact Assessment, 2005b). The
“rule of three” protects confidentiality by prohibiting the reporting of information that might be
attributed to a single business or individual. This keeps potential competitors from gaining
inside information about the activities of that business or individual (Impact Assessment, 2005b).
There are many small rural communities that have only one or two fish processors that contribute
a relatively large amount of jobs and income to the local economy (Impact Assessment, 2005b).
Nonetheless the data cannot be reported because of the rule of three. In many cases this
essentially makes community-level landings data unavailable to researchers outside of NMFS
because of the sensitive and confidential nature of the information. However, since our results
33
are standardized to reflect landings per 1,000 residents it would not violate federal confidentiality
rules.
(B) Ground-Truthing
(1)
Ethnographic Field Research
(a) Field Research Plans and Sites
Field work was conducted from June 23, 2008 until August 7, 2008 in the following Texas
communities: Bacliff, Galveston, Kemah/Seabrook (treated as one community), San Leon, and
Texas City, all situated on Galveston Bay; Port O’Connor, Seadrift and Port Lavaca situated on
San Antonio Bay; and Palacios situated on Matagorda Bay (Figure 1).
Figure 1: Map of Galveston, Matagorda, and San Antonio Bays.
34
The initial goal of the field work was to construct in-depth community profiles for two
communities, one in Galveston Bay and the other in San Antonio Bay. These bay systems were
chosen because they are at different stages of gentrification and urbanization. The Galveston Bay
Complex has been undergoing rapid and profound change for the past 30 years, and massive
gentrification has already occurred. In contrast, the gentrification of San Antonio Bay is more
recent, and some communities are just now beginning to see profound changes. Thus, the
communities selected for study will have experienced very different rates of level one (chronic)
perturbation.
As planning progressed, the decision was made to carry out field research in a larger number of
communities, specifically in nine communities on three bay systems. The reorientation was
made for several reasons. Galveston Bay fishermen no longer live in the communities in which
they fish, and in order to capture an array of interviewees, coverage in this region was expanded
to four communities. Additionally, the communities around Galveston Bay are no longer
geographically and socially distinct entities, even if they are politically separate. To illustrate
how compacted the communities are, one encounters six different municipalities along an
approximately 10 mile strip on NASA Road 1 – Houston, Webster, El Lago, Nassau Bay,
Pasadena, and Seabrook – without seeing any distinct changes between communities. Thus, the
decision was made to treat the western shore of Galveston Bay from Galveston Island to
Seabrook as one fishing region ethnographically.
The Galveston Bay Complex (referred to subsequently as Galveston Bay) consists of the
interconnected Galveston, East, West, Trinity and Chocolate Bays. It is situated southeast of
Houston, the 4th largest city in the nation with a population of over 2 million, and the
metropolitan population is approximately 4.5 million. The southeastern city limit is only a few
miles from the shoreline. The four geographical entities of the region that carry the name
Galveston are: the barrier island (Galveston Island) that separates the Gulf of Mexico from the
bay; the bay itself; the city that occupies Galveston Island; and the county that covers both the
island and much of the western mainland shore of the bay. With the exception of the city of
Galveston, the communities studied are situated on the mainland, on the western shore of the
bay. The area is shown below in Figure 2.
35
Figure 2: Map of the Galveston Bay geographical entities.
In addition to being one of the largest metropolitan areas in the United States, the area also is
heavily industrialized. Large petrochemical complexes and oil fields are situated in the cities of
La Porte, Deer Park and Pasadena to the northeast of the bay and along the southeastern coast in
Texas City. A ship channel, some 200 feet wide and 40 feet deep, cuts across the bay to allow
ships access to the Port of Houston, the largest port in the Gulf of Mexico and one of the largest
in the United States.
The southern portion of the research area consists of the communities on the shores of a series of
connected bays: San Antonio Bay, Matagorda Bay, Espiritu Santu Bay and Lavaca Bay.
Although the communities surrounding San Antonio, Espiritu and Lavaca Bays (Port O’Connor,
Seadrift and Port Lavaca) are discrete geographically, separated by farms and ranches, there are
considerable links and socioeconomic interaction between fishermen, fish houses, and processors
in these communities. The decline in fishing infrastructure, as detailed below, has meant that
fishermen must go to other towns to sell their catch and in some cases, to dock. Thus
respondents would also direct us to other interviewees in nearby towns. Due to our ability to
extend our time in the field (thanks to TPWD’s generosity with housing), the decision was made
36
to add Palacios to the study, which is on Matagorda Bay. Palacios has a large Gulf fleet and
significant Vietnamese and Hispanic fishers.
The landscape surrounding the San Antonio - Espiritu Santos Bay Complex and the Matagorda Lavaca Bay Complex differs significantly from that of the Galveston Bay Complex. The largest
metropolitan entity is Victoria, a city of approximately 60,000, situated about an hour to the
northwest. Other towns in the region have populations of less than 12,000, most below 3,000
and are geographically bounded by producing farms and ranches. Although the communities in
the study, Palacios, Port Lavaca, Seadrift, and Port O’Connor are relatively small towns, they
differ considerably in socioeconomic characteristics and in terms of their historic and current
dependence on fisheries, as will be discussed in more detail below. The region is shown in
Figure 3, with San Antonio Bay to the southwest of Matagorda Bay. The community of Port
O’Connor is at the southern-most point of the peninsula, below Seadrift.
Figure 3: Map of San Antonio and Matagorda Bays.
37
(b) Ethnographic Interview Protocol
An interview protocol was developed by the HARC Post-doctoral Fellow in Human Natural
Systems, under Dr. Weeks’ guidance and modified and refined with Dr. Blount’s guidance at a
workshop held at HARC on June 13, 2008. Dr. Blount coordinated the changes and oversaw the
final draft of the protocol. The final copy was a semi-structured interview protocol, containing
16 open-ended questions. The objective was for the intern field researchers to ask respondents
the same questions, thereby providing comparable answers but also allowing respondents to
elaborate freely in their responses. A copy of the interview questionnaire is attached as
Appendix A.
(c) Field Research Methods and Data Collection
As a summary overview, semi-structured interviews were conducted with agency staff,
fishermen, community leaders and business owners. Fishermen held an array of license types
including crab, bay, bait and gulf shrimp, oysters and finfish. Additionally, informal
conversations and short encounters on docks and in fish houses added context to the formal
interviews. All encounters were noted and assigned identifying numbers. TPWD staff
introduced researchers to a few key informants in each bay system. More interviewees were
identified using a chained-referral technique, but several respondents were recruited in random
encounters on docks, in fish houses, and offices. Historical information on each community was
collected in order to situate current changes in the fishery and in the communities as a whole.
Infrastructure was noted using GPS and photographic data.
(i) Galveston and Galveston Bay Communities
The field research, again, was guided by the interview protocol developed in June 2008, with one
exception. In addition to conducting interviews with fishermen, information on gentrification
was collected in two ways: a) interviews with developers, bankers, elected officials and b) the
collection of written materials about proposed developments. While the research methods were
historic and ethnographic, the inventory of questions was different. The gentrification research
in Galveston was carried by one of the four interns, Jerry Lord, a doctoral student in
anthropology at the University of Texas who specializes in gentrification. His field research
was supervised by Dr. Weeks.
Field research with fishermen in the Galveston Bay communities of Bacliff, Galveston, Kemah/
Seabrook, San Leon, and Texas City was carried out by Meredith Marchioni, a doctoral student
in anthropology at Florida International University who specializes in fisheries. Her research
was supervised by Dr. Weeks. The research in the four more southern communities was
completed by another intern, Beth Croucher, a Masters graduate in anthropology at the
University of Denver. She conducted the field studies in Palacios and Port O’Connor, while Dr.
Lovette Miller, a recent doctoral graduate in geography from the University of Maryland and a
HARC Post-doctoral Fellow in Human and Natural Systems, carried out the research in Seadrift
38
and Port Lavaca. The field research in the Matagorda Bay and San Antonio Bay communities
was supervised by Dr. Weeks and Dr. Blount.
Interviews proved to be considerably more difficult to secure in the Galveston Bay communities
than in the communities in the other two bays. The difficulty appeared to stem from the fact that
fishers were urban dwellers who typically did not live in the communities where they fished.
Interviews were by necessity conducted mostly as dock intercepts (with some chain referral from
PISCES, the local fishermen’s organization and Texas Parks and Wildlife Department) which
affected interview content. In that context, the fishers’ responses tended to be to the point, even
terse, and with little elaboration. Follow-up “probe’ questions were difficult to do, and when
asked tended not to be successful.
Given the limitations with interviews in the Galveston Bay Complex, the decision was made to
collapse the interviews into one data set for all of the communities. Differences across the five
communities are thus not distinguished. There was an average of only approximately seven
interviews per community, however, and the information per community was thus limited in
comparison with the communities in San Antonio and Matagorda Bays. It should thus be noted
that the scores for the Galveston Bay communities will likely be depressed substantially in
comparison with the communities in the more southern bays, a function of the lesser interview
content.
Despite the loss of information from collapsing the limited data sets in the Galveston Bay
Complex, the communities are all in spatial and demographic contexts that are more similar to
each other than is the case in the other four communities. The communities are all in a suburban,
high population density area, essentially part of the Houston metro-complex. To help offset the
loss of differential information across communities, however, the second type of ground-truthing
analysis, comparisons across historical and background community context, treats each of the
nine communities individually.
In addition to the problem of getting detailed interviews, Hurricane Ike struck the Galveston Bay
communities on 13 September 2008, severely altering the conditions of the communities and
preventing any follow-up or continuation of interviews. The impacts of Hurricane Ike will be
discussed in Appendix E.
(d) Number of Interviews
The number of interviews completed in the Matagorda Bay and San Antonio Bay communities is
shown in Table 13.
39
Table 13: Number of interviews completed in the coastal communities.
COMMUNITY
NUMBER OF COMPLETED INTERVIEWS
Galveston Bay
Port Lavaca
Seadrift
Palacios
Port O’Connor
36
17
20
13
20
TOTAL
106
(e) Coding Procedure
A coding procedure was developed as a means of analyzing the interviews. Drs. Blount, Weeks,
and Miller compiled a list of topics and their keyword descriptors independently from a selected
sub-set of the interviews, 15 all together. Each researcher also spot-checked coding across the
communities to promote consistency. The three researchers compared their results, several
times, each producing modifications and refinements. The communications about the coding
were by email and attachments, but also by a conference call, leading to the development of a
final coding list and sheet.
Approximately 60 items were coded, and for convenience of analysis, keywords on related topics
were subsumed under descriptive headings. All comments/keywords on decline in number of
fishers, in number of boats, in number of employees, for example, were listed under a heading
Infrastructure.
The procedure for coding was to read through each interview and to record on a coding sheet
whether the respondent mentioned or referred to one of the topics, by keywords (see Blount,
2002; Blount and Kitner, 2007). The procedure is often used in cognitive studies, the assumption
being that the more salient or important a topic is to an individual, the more likely the topic will
be mentioned and discussed. Failure on the part of a respondent to mention a term does not
imply that the topic is unknown or unimportant, but that is taken as an indication of relative
lesser importance. Only one instance per interview was recorded, to preclude multiple
references to the topic in a stretch of a response to one question. The coding for Galveston Bay
was completed by Dr. Weeks, the coding for Port Lavaca and Seadrift was completed by Dr.
Miller, and the coding for Palacios and Port O’Connor was coded by Dr. Blount. A copy of the
coding sheet and the total number of topics/codes by community can be seen in Table 28.
(f) Features of Data Analysis
Analyses of the code data was conducted by Dr. Blount. Total scores were calculated for each
topic/code by community, as reported in Table 28, and comparisons were made across
communities on the basis of the totals. Given that the total number of interviews across the
40
communities differed, the total raw scores per community were then divided by the number of
interviews, to provide a standard rate measure. Since those calculations produced a high number
of decimal fractions, the scores were multiplied by 100, as shown in Table 29. Comparisons
were made of the rate score totals by community.
In addition to the distribution of rate scores, the topics/codes were assessed as contributing to
fishery vulnerability or fishery resilience. The scores on the topics/codes in each of those
categories were totaled, and the communities were ranked comparatively in terms of the highest
to lowest vulnerability and resilience.
(2) Compilation of Historical and Contextual Information
Dr. Weeks assumed primary responsibility for this section of ground-truthing. Information was
collected from brief histories of the communities and gentrification compiled by the interns. In
addition, information was used from HARC databases and from relevant publications.
Information was collected on economic and social dependence of the nine communities on
commercial and recreational fishing, reported in Table 38 and Table 39, and on community
socioeconomic vulnerability, reported in Table 40.
Data were also accumulated on
gentrification, as reported in Table 41.
Dr. Weeks also had primary responsibility for carrying out the historical analyses of the
communities, as reported in Appendix B. Jerry Lord compiled the information under Dr. Weeks’
supervision on gentrification collected by the field researchers, as reported in Appendix C and D,
and for the impacts of Hurricane Ike on the Galveston Bay communities, as reported in Appendix
E.
3.
Project Personnel
Principal Investigator:
Ms. Judy L. Jamison
Foundation Staff:
Dr. Michael Jepson
Mr. Frank C. Helies
Ms. Gwen Hughes
Ms. Charlotte Irsch
Independent Contractors:
Dr. Benjamin Blount
Dr. Priscilla Weeks
Dr. Steve Jacob
Executive Director
Program Director (former)
Program Director (current)
Program Specialist
Grants/Contracts Specialist
Administrative Assistant
University of Texas, San Antonio
HARC
York College
41
Overall project quality control and assurance was assumed by the Gulf & South Atlantic
Fisheries Foundation, Inc. through its office in Tampa, FL. The Foundation’s Executive Director
had ultimate responsibility for all Foundation administrative and programmatic activities, with
oversight by the Foundation’s Board of Trustees. She ensured timely progress of activities to
meet project objectives and confirmed compliance of all activities with NOAA/NMFS. The
Foundation’s Program Directors had overall responsibility for all technical aspects of Foundation
projects and coordinated performance activities of all project personnel, including contractors.
The Program Directors prepared all progress reports concerning project performance.
It was the responsibility of the Foundation’s Executive and Program Directors to ensure that
quality control and assurance were maintained for all aspects of this program. This was
accomplished through regular phone and email communications with project Contractors.
The Program Specialist was responsible for tracking programmatic activities, including
generating supporting documentation to assist in any and all programmatic audits. She organized
workshops and was responsible for auditing and paying all program related invoices. She
processed requests for reimbursement to conform with federal guidelines and prepared and
maintained all subcontracts and amendments.
The Grant/Contracts Specialist was responsible for maintaining general financial accounting of
all Foundation funds including all Cooperative Agreements and contracts, as well as
communicating with NOAA Grants Management personnel, and assisting fiscal auditors in their
reviews. She conducted/documented internal and program (single and desk) audits, prepared
backup documentation for fiscal audits, and drafted award extension requests (as applicable).
She provided the Executive and Program Directors with projected budgets concerning program
performance and ensured that these budgets adhered to the proposed budget. Finally, she
prepared the annual administrative budget, NOAA Financial Reports, and confirmed compliance
of all activities with NOAA/NMFS and OMB guidelines.
The Administrative Assistant was responsible for receptionist/clerical duties, word processing,
filing correspondence, dissemination of materials to industry (final reports, press releases,
newsletters). She was also responsible for creating and organizing meeting files, processing
invoices and maintaining cooperative program files.
This project required the cooperation and active participation of individuals with expertise and
research experience within the field of community studies and indicators research. The
Foundation contracted with several individuals in conjunction with this project who then hired
research assistants with the appropriate background for this research. These essential personnel
needed to complete project objectives are:
Dr. Ben Blount’s human ecology interests have been focused on coastal communities to examine
the knowledge systems of fishing communities, especially to the problems that the communities
see themselves facing for livelihood survival. Human ecology includes knowledge of the options
42
open to individuals in relation to ecological resources and the consequences of decision making.
Dr. Blount’s research into those topics has been carried out over the past decade among fishing
communities of the Atlantic coast of Georgia and North Carolina.
Dr. Priscilla Weeks is an Environmental Anthropologist in the Social and Policy Analysis Group
at Houston Advanced Research Center. Her research interests include: the public understanding
and acceptance of the scientific models that inform resource management; the social impacts of
technical and resource management innovations; and the way in which scientific information,
cultural models and values combine in environmental disputes. Recent research has examined
issues related to protected areas and fisheries. Dr. Weeks has worked to incorporate local
knowledge into environmental management through collaborative decision making and has
served as a facilitator for both state and federal agencies. She has conducted research along the
Texas coast investigating issues within several different fisheries and is familiar with many
communities along its coast.
Dr. Steve Jacob is a community sociologist who specializes in natural resource issues. He has
had extensive experience conducting community case studies that have incorporated both
primary and secondary data and qualitative and quantitative methods. Much of Dr. Jacob’s
research has focused upon coastal communities in Florida and on forested communities in
Pennsylvania.
VI. FINDINGS
A. Actual Accomplishments and Findings
1. Social Indicators Indices
The research findings of the social indicators are presented below, in text and tabular form.
(A) Urban Sprawl Index
The urban sprawl index (Table 14) consists of seven items: 1) distance to the nearest city with a
population of 50,000 or greater (range 0 to 54.9 miles), 2) percentage population density change
2000 to 2007 (range -5 to 36 percent), 3) percentage of homes built between 1995 and 1998
(range 1.5 to 16.5 percent), 4) percentage of homes built between 1999 and 2000 (range 1.1 to 26
percent), 5) percentage of population who lived in another county in 1995 (range 9.1 to 37.1
percent), 6) percentage of homes valued at less than $100,000 in 2000 (range 6.9 to 99.2
percent), and last 7) cost of living index with an average for the U.S. set at 100 (range 73.8 to
95.6).
43
Table 14: The Urban Sprawl Index.
Community
Nearest
city w/
50k pop
in miles
Population
density
change
2000-2007
Percentage
homes built
between
1995-1998
Percentage
homes built
between
1999-2000
Percentage
population
who lived in
another
county
1995
Percentage
of homes
LT 100K
2000
Cost of
living
index 2008
USA avg =
100
Urban
Sprawl
Factor
Score
Ranking
Port Lavaca
Sea Drift
Port O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Little River
Mount Pleasant
28.6
34
34
53.8
6.5
14.6
0
11.1
11.4
54.9
0
-0.05
0.06
0.15
-0.02
0.22
0.13
-0.01
0.01
0.13
0.26
0.36
5
6.9
8.3
4.9
16.5
7.4
3.3
1.5
8.4
5.6
15.9
1.1
2.2
7.9
1.7
3.3
3.1
1.1
1.9
2.4
26
8.4
12.4
20
20.8
18.4
31.9
27.2
22.2
9.1
26.8
37.1
28.4
86.3
99.2
80.2
87.5
36.9
68
7.4
84.2
75.4
26.5
6.9
75.6
73.8
82.1
75.4
88.8
86.8
89.4
85.5
84.8
91.2
95.6
-0.862
-0.744
0.101
-0.843
0.401
-0.010
-0.502
-0.823
-0.201
1.771
1.689
11
8
4
10
3
5
7
9
6
1
2
0.834
0.788
0.744
-0.711
0.471
PC components
Factor Loading
0.354
0.617
Theta
0.787
Eigenvalue
3.077
Percentage Explained Variation
43.950
Single Factor Solution
Higher Ranking = Closest to Sprawl
Lower Ranking = Distant from Sprawl
The principal components analysis for the above items in the urban sprawl index produced a
single factor solution suggesting that the component variables only measure one underlying
construct. The eigenvalue for the single factor solution was 3.077 which explained 43.95 percent
of the variation in the model. The Armor’s Theta Reliability coefficient for the index was 0.787
indicating an adequate level of internal consistency for a seven item index. The factor loadings
can be seen for each variable underneath the appropriate variable column. The factor loadings
ranged from a high of 0.834 to a low of 0.354. The strongest loadings were for the percentage of
homes built between 1995 and 1998 (0.834), the percentage of homes built between 1999 and
2000 (0.788), and percentage population who lived in another county in 1995 (0.744). All factor
loadings were above 0.350 and so were included in the scale.
Variables 1, 2, and 7 were downloaded from the website “City-Data.com.” Variables 3, 4, 5, and
6 were derived from the 2000 Decennial Census Summary Tape File 3. Variable 2 was
44
calculated by subtracting the year 2000 density from the 2007 density and then dividing by the
2000 density. Last the result is multiplied by 100 to produce a percentage.
(B) Natural Resource Migration Index
The natural resource migration index (Table 15) consisted of five variables: 1) the percentage of
homes in 2000 that were rented (range 15.3 to 56.4 percent), 2) The percentage of homes that
were vacant in 2000 (range 5.8 to 63.2 percent), 3)the number of boat ramps within the
municipality (range of 1 to 12), 4) the percentage of water cover within the municipal boundaries
(range 0 to 77.8 percent), and last 5) the percentage of the workforce employed in agriculture,
fishing, or mining (range 1.24 to 17.7 percent).
Table 15: The Natural Resource Migration Index.
Community
Percentage
homes
rented
2000
Percentage
homes
vacant
2000
Number of
boat ramps
Percentage
water cover
Ag, farming
fishing,
mining % in
industry
Resource
Migration
Score
Ranking
Port Lavaca
Sea Drift
Port O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Little River
Mount Pleasant
34.5
20.7
15.3
30.8
48.1
33.8
56.4
34.1
33.3
18.2
26
12.6
25.2
63.2
15.9
9.7
9
20.6
7.2
14.8
30.3
5.8
5
3
5
1
3
4
12
4
2
3
2
28.7
0
25
3.7
73.4
5.8
77.8
62.8
0
3.7
15.35
3.61
17.7
20
10.05
2.45
5.12
1.53
1.24
2.37
4.18
1.48
1.536
-1.336
0.185
-1.865
1.489
-0.398
6.608
1.788
-1.332
-0.110
1.278
3
10
6
11
4
8
1
2
9
7
5
0.471
0.616
0.791
-0.357
PC components
Factor Loading
0.865
Theta Reliability
0.701
Eigenvalue
2.270
Percentage Explained Variation
43.071
Single Factor Solution
High Ranking = High Migration
Low Ranking = Low Migration
The principal components analysis for these items produced a single factor solution with an
eigenvalue of 2.27 and an explained variation of 43.07 percent. The Armor’s Theta Reliability
coefficient for this index is 0.701 reflecting adequate levels of internal consistency for the index.
The factor loadings for the variables in the index ranged from a high of 0.865 to a low of -0.357.
The strongest loadings were for the percentage of homes rented (0.865), percentage of water
cover in the community boundaries (0.791), and the number of boat ramps in the community
boundaries (0.616). All factor loadings were greater than 0.350 and were included in the index.
45
Variables 1, 2, and 5 were derived from the 2000 Decennial Census Summary Tape File 3.
Variable 3 was from a custom NOAA NMFS database. Variable 4 was derived from Census
data by dividing total land area in the municipality into the percentage water cover.
(C) Retirement Migration Index
The retirement migration index (Table 16) consists of four items: 1) the percentage of the
population over age 65 (range 5.7 to 23.4 percent), 2) the percentage population receiving social
security (range 13.3 to 36.4 percent), 3) the mean retirement income (range $12,305 to $24,264),
and 4) the percentage of the labor force working in services (range 9.5 to 24.2 percent).
The principal components analysis produced a single factor solution and an eigenvalue of 2.681
and an explained variation of 67.03 percent. The Armor’s Theta Reliability coefficient for this
index was 0.783 reflecting an adequate level of internal consistency. The factor loadings ranged
from a high of 0.934 to a low of 0.421. The strongest loadings were for the variable percentage
population over age 65 (0.934), the percentage population receiving social security (0.929), and
the mean retirement income in the community (0.877). All loadings were above 0.350 and were
included in the index.
Table 16: The Retirement Migration Index.
Community
Port Lavaca
Sea Drift
Port O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Little River
Mount Pleasant
Percentage
population
over age 65
Percentage
population
receiving
social security
Mean
Retirement
income
Percentage labor
force working in
services
Retiree
Migration
Score
Ranking
12.1
13.5
19.6
11.1
5.7
10.3
13.7
13.5
8.2
23.4
10.3
27.5
36
36.4
31.5
13.3
21.8
25.8
28.4
18.2
41.8
18.2
$20,122
$12,305
$18,771
$24,264
$19,662
$16,049
$20,469
$20,235
$12,375
$16,834
$22,961
15.9
15.6
19.1
9.5
10.3
14.4
24.2
13.7
19.3
19.2
11.8
0.121
0.801
1.387
0.382
-1.338
-0.292
0.198
0.531
-0.831
2.861
-0.263
7
3
2
5
11
9
6
4
10
1
8
0.929
0.877
0.42
PC components
Factor Loading
0.934
Theta Reliability
0.783
Eigenvalue
2.681
Percentage Explained Variation
67.03
Single Factor Solution
High Ranking = High Migration
Low Ranking = Low Migration
Variables 1, 2, 3, and 4 were derived from the 2000 Decennial Census Summary Tape File 3.
For variables 1 and 2 the percentage population over age 65 and receiving social security were
46
calculated by taking the absolute number of those 65 and older and receiving social security and
dividing it by the total population in the community.
(D) Population Composition Vulnerability/Resiliency Index
The population composition vulnerability/resiliency index (Table 17) consisted of seven
variables: 1) percentage of whites in the community (range 57.1 to 91.4 percent), 2) percentage
of female singled headed households (range 7.7 to 16.9 percent), 3) all parents in the household
are in workforce with children under six years old (range 23.3 to 64.7 percent), 4) percentage
that speak a language other than English in the home (range 3.7 to 49.9 percent), 5) percentage
population less than 18 and greater than 65 (range 29.6 to 46.5 percent), 6) the percentage of
high school graduates (range 58.2 to 92.7 percent), and 7) the percentage of college graduates
(6.7 to 52.6 percent).
Table 17: The Population Composition Vulnerability/Resiliency Index.
Community
Port Lavaca
Seadrift
Port O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Little River
Mount Pleasant
Percent
whites
2000
Percent
female
single
headed
HH
71.9
77.7
90.4
57.1
88.9
80.4
58.7
60.3
82.6
91.4
90.2
13.7
9.2
7.7
13.4
8.1
7.3
16.9
7.4
12
8.4
8.3
Both
parents
in work
force w/
kids > 6
years
55.4
46.7
45.2
23.3
56.2
32.1
55.7
41.2
47.8
64.7
51.1
-0.408
0.661
PC components
Factor Loading
0.715
Theta Reliability
0.755
Eigenvalue
2.698
Percentage Explained Variation
44.974
Percent
speak
other
language
2000
Percent >
18 and <
65 2000
Percent
high
school
2000
40.3
32.4
22.4
49.9
14.6
22.4
26.5
15.4
20.6
3.7
6.2
42.4
36.3
45.5
46.5
29.6
33.3
37.1
42.2
37.2
39.1
35.4
68.1
58.2
70.6
57.2
92.7
76
74.4
77.1
69.9
89.2
94
-0.728
-0.486
0.944
Single Factor Solution
Percent Population
college Composition
Score
17.7
6.7
11.8
12.6
41.2
11.9
23.7
11.2
8.6
21.4
52.6
-1.066
-1.238
-0.455
-1.923
1.211
-0.048
-0.505
-0.487
-0.322
0.627
1.379
9
10
6
11
2
4
8
7
5
3
1
0.733
High Ranking = More Resilient
Low Ranking = More Vulnerable
The principal components analysis produced a single factor solution with an eigenvalue of 2.698
with an explained variation of 44.974 percent. The Armor’s Theta Reliability coefficient for this
index was 0.755, which represents adequate levels of internal consistency. The factor loadings
ranged from -0.408 to 0.944. The strongest loadings were for the variables percentage high
47
Ranking
school graduates (0.944), percentage college graduates (0.733), and percentage whites (0.715).
All the factor loadings were above 0.350 and so were included in the index.
All variables in this index were from the 2000 Decennial Census Summary Tape File 3.
Variable 5 was calculated by adding the absolute number of residents under 18 and over 65
together and then dividing by the total population to derived a percentage.
(E) Poverty Index
The poverty index (Table 18) incorporated five items into an index: 1) the percentage population
in poverty in 2007 (range 5.1 to 25.1 percent), 2) the percentage population 50% below the
poverty line in 2007 (range 2.7 to 13.5 percent), 3) the percentage of population over 65 in
poverty in 2000 (range 1.7 to 16.3 percent), 4) the percentage under 18 in poverty in 2000 (range
5.3 to 33.5 percent), and 5) the cost of living index for 2008 (range 73.8 to 95.6).
Table 18: The Poverty Index.
Community
Port Lavaca
Sea Drift
Port O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Little River
Mount Pleasant
Percentage
population
in poverty
2007
Percentage
population
50%
under line
2007
Percentage
over 65 in
poverty
2000
Percentage
under 18
in poverty
2000
Cost of
living index
2008 USA
avg = 100
Poverty
Index
Score
Ranking
20.1
25.1
6.8
24.2
5.5
19.7
22.3
15.2
21.7
7.5
5.1
8.6
13.5
4.6
9.9
3
11.9
10.5
4.2
8.9
3.7
2.7
14.7
11.7
11.1
14.3
5.1
2.3
14.2
11.3
16.3
1.7
6.6
25
33.1
19.7
28.9
6.1
33.5
32.1
19.3
30.6
15.3
5.3
75.6
73.8
82.1
75.4
88.8
86.8
89.4
85.5
84.8
91.2
95.6
1.221
1.929
0.147
1.563
-0.680
1.090
1.347
0.324
1.322
-0.485
-0.848
5
1
8
2
10
6
3
7
4
9
11
0.808
0.746
0.945
-0.601
PC components
Factor Loading
0.974
Theta Reliability
0.900
Eigenvalue
3.581
Percentage Explained Variation
71.625
Single Factor Solution
High Ranking = More Vulnerable
Low Ranking = More Resilient
The principal components analysis produced a single factor solution with an eigenvalue of 3.581.
The explained variation for the model was 71.625 percent. The Armor’s Theta Reliability
coefficient for the index was 0.900 which represents high levels of internal consistency for this
index. The factor loadings from the principal component analysis ranged from -0.601 to 0.974.
48
The strongest loadings in the analysis were for the percentage population in poverty in 2007
(0.974), the percentage under 18 in poverty in 2000 (0.945), and the percentage population 50%
below the poverty line in 2007 (0.808). All the factor loadings were above 0.350 and so were
included in the index.
Variables 1, 2, and 5 were downloaded from the website “City-Data.com.” Variables 3 and 4 are
from the 2000 Decennial Census Summary Tape File 3.
(F) Housing Characteristics Vulnerability/Resiliency Index
The housing characteristics vulnerability/resiliency index (Table 19) consisted of five items: 1)
median rent in dollars in 2000 (range $430 to $832), 2) median mortgage in dollars in 2000
(range $546 to $1,302), 3) median number of rooms in houses (range 4.1 to 6.2 rooms), 4) the
percentage of houses with inadequate plumbing (range 0.1 to 1.7 percent), and 5) house age in
years 2008 (range 18 to 54 years).
Table 19: The Housing Characteristics Vulnerability/Resiliency Index.
Community
Percentage
change
Median
Rent 2000
Percentage
change
median
mortgage
2000
Percentage
change in
property
values
2007
Percentage
change in
# of
renters
2000
Percentage
moved
into house
1999-2000
Percentage
moved
into house
1995-1998
Housing
Disruptions
Score
Ranking
Port Lavaca
Sea Drift
Port O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Little River
Mount Pleasant
29.1
47.7
-6.1
19.2
46.7
30.3
39.4
32.2
34.0
33.9
54.9
53.0
32.2
92.3
74.3
64.2
41.0
24.0
48.9
31.2
38.0
47.3
52.3
80.4
42.7
49.4
56.4
66.4
66.4
66.2
66.4
90.6
107.5
-5.7
-22.9
-31.5
-6.7
-41.2
48.0
-2.8
-7.1
-2.7
-43.7
-31.5
19.8
19.7
15.5
20.5
33.6
21.1
27.6
19.4
27.5
22.6
28.7
29.4
30.3
29.9
32
35.3
29.8
29.2
24.3
29.2
39.8
35.3
-0.495
0.326
-1.746
-0.866
0.895
-0.920
0.136
-0.764
0.173
1.111
1.387
7
4
11
9
3
10
6
8
5
2
1
-0.543
0.792
-0.486
0.728
0.67
PC components
Factor Loading
0.865
Theta
0.789
Eigenvalue
3.088
Percentage Explained Variation
44.108
Single Factor Solution
Low Ranking = More Vulnerable
High Ranking = More Resilient
The principal component analysis produced a single factor solution with an eigenvalue of 2.592
and an explained variation of 51.937 percent. The Armor’s Theta Reliability coefficient for this
index was 0.768 reflecting adequate levels of internal consistency. The factor loadings ranged
from -0.361 to 0.884. The strongest loadings were for the variables median mortgage costs in
49
dollars (0.884), median number of rooms per house (0.880), and median rent costs in dollars
(0.841). All the factor loadings were above 0.350 and so were included in the index. Variables 1
through 5 were from the 2000 Decennial Census Summary Tape File 3.
(G) Labor Force Structure Index
The labor force structure index (Table 20) consists of five variables 1) median household income
in 2007 (range $35,610 to $77,349), 2) percent unemployed in 2007 (range 2.2 to 11.4 percent),
3) percent population in the labor force in 2000 (range 48.3 to 75.9 percent), 4) worker
classification percent self employed in 2000 (5.6 to 26 percent), and 5) percent of population
receiving supplemental security income (disability) (range 1.6 to 8.7 percent).
Table 20: The Labor Force Structure Index.
Community
Port Lavaca
Sea Drift
Port O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Little River
Mount Pleasant
Median
Household
Income 2007
Percent
unemployed
2007
Percent
population in
the labor
force 2000
Class of
Worker Self
Employed
2000
Percent
people
receiving SSI
2000
Economic
Structure
Score
Ranking
$39,199
$36,742
$47,854
$31,623
$63,507
$39,051
$35,610
$38,924
$39,668
$47,869
$77,349
4.2
5.8
5.0
11.4
3.0
11.4
10.1
8.1
5.3
3.4
2.2
58.8
48.3
51.0
50.7
75.9
62.5
59.7
58.4
66.5
58.0
69.9
6.7
8.5
26.0
8.8
6.6
11.5
5.6
9.2
8.5
11.1
7.9
5.7
5.4
6.4
8.7
1.8
4.0
5.3
5.3
3.8
1.6
2.1
-0.925
-0.871
-0.833
-1.618
0.628
-0.671
-1.001
-0.802
-0.513
0.368
0.907
9
8
7
11
3
5
10
6
4
2
1
-0.672
0.351
0.741
-0.702
PC components
Factor Loading
0.883
Theta Reliability
0.727
Eigenvalue
2.392
Percentage Explained Variation
47.834
Single Factor Solution
High Ranking = More Resilient
Low Ranking = More Vulnerable
The principal component analysis produced a single factor solution with an eigenvalue of 2.392
and an explained variation of 47.834 percent. The Armor’s Theta Reliability coefficient for this
index was 0.727 reflecting adequate levels of internal consistency. The factor loadings for this
analysis ranged from 0.351 to 0.883. The strongest loading in the analysis were for the variables
median household income (0.883), percentage of self employed workers (0.741), and percent
people receiving supplemental security income (disability) (-0.702). All the factor loadings were
above 0.350 and so were included in the index.
50
Variables 1 and 2 were downloaded from the website “City-Data.com.” Variables 3,4, and 5
were from the 2000 Decennial Census Summary Tape File 3.
(H) Natural and Technological Disaster Risk Index
The natural and technological disaster risk index (Table 21) consists of five variables: 1)
standardized (U.S. average = 100) damaging hail risk (range 40 to 154), 2) standardized (U.S.
average = 100) damaging hurricanes risk (range 355 to 499), 3) standardized (U.S. average =
100) damaging tornadoes (range 45 to 473), 4) standardized (U.S. average = 100) damaging
winds (range 13 to 133), and 5) Environmental Protection Agency Registered Facilities (range 2
to 212).
Table 21: The Natural and Technological Disaster Risk Index.
Community
Port Lavaca
Sea Drift
Port O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Little River
Mount Pleasant
Damaging
Hail Risk
Average
USA AVG =
100
Damaging
Hurricanes
Risk Average
USA AVG =
100
Damaging
Tornadoes
Risk Average
USA AVG =
100
Damaging
Winds Risk
Average
USA AVG =
100
EPA
Registered
facilities
Potential
Disaster
Score
Ranking
77
57
57
76
130
132
100
140
154
40
90
355
371
371
398
442
465
499
488
458
431
436
110
45
45
104
468
429
265
349
473
35
60
15
13
13
33
92
92
69
127
105
35
133
20
2
2
5
8
6
189
212
8
16
84
-1.237
-1.561
-1.581
-0.894
1.241
1.549
0.982
1.824
1.666
-1.217
0.409
9
11
10
7
4
3
5
1
2
8
6
0.868
0.579
0.727
0.368
PC components
Factor Loading
0.867
Theta Reliability
0.869
Eigenvalue
3.232
Percentage Explained Variation
64.632
Single Factor Solution
High Ranking = More Vulnerable
Low Ranking = Less Vulnerable
The principal components analysis produced a single factor solution with an eigenvalue of 3.232
and an explained variation of 64.632 percent. The Armor’s Theta coefficient for this index was
0.869 reflecting high levels of internal consistency. The factor loadings for this analysis ranged
from 0.368 to 0.868. The strongest loadings were for the variables damaging hurricanes (0.868),
damaging hail (0.867), and damaging winds (0.727). All the factor loadings were above 0.350
and so were included in the index.
51
Variables 1 through 4 were downloaded from “Moving.com.” Variable five was derived from
the EPA’s “EnviroMapper” website.
(I) Housing Disruptions Index
The housing disruptions index (Table 22) consists of six variables: 1) percentage change in
median rent 1990 to 2000 (range -6.1 to 54.9 percent), 2) percentage change in median mortgage
costs 1990 to 2000 (range 24 to 92.3 percent), 3) percentage change in home values 2000 to
2007(range 42.7 to 107.5 percent), 4) percentage change in number of renters 1990 to 2000 (43.7 to 48 percent), 5) percentage of residents who moved into their current house in 1999-2000
(range 15.5 to 33.6 percent), and 6) percentage of residents who moved into their current house
in 1995-1998 (range 24.3 to 39.8 percent).
Table 22: The Housing Disruptions Index.
Community
Percentage
change
Median
Rent 2000
Percentage
change
median
mortgage
2000
Percentage
change in
property
values
2007
Percentage
change in
# of
renters
2000
Percentage
moved
into house
1999-2000
Percentage
moved
into house
1995-1998
Housing
Disruptions
Score
Ranking
Port Lavaca
Sea Drift
Port O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Little River
Mount Pleasant
29.1
47.7
-6.1
19.2
46.7
30.3
39.4
32.2
34.0
33.9
54.9
53.0
32.2
92.3
74.3
64.2
41.0
24.0
48.9
31.2
38.0
47.3
52.3
80.4
42.7
49.4
56.4
66.4
66.4
66.2
66.4
90.6
107.5
-5.7
-22.9
-31.5
-6.7
-41.2
48.0
-2.8
-7.1
-2.7
-43.7
-31.5
19.8
19.7
15.5
20.5
33.6
21.1
27.6
19.4
27.5
22.6
28.7
29.4
30.3
29.9
32.0
35.3
29.8
29.2
24.3
29.2
39.8
35.3
-0.495
0.326
-1.746
-0.866
0.895
-0.920
0.136
-0.764
0.173
1.111
1.387
7
4
11
9
3
10
6
8
5
2
1
-0.543
0.792
-0.486
0.728
0.67
PC components
Factor Loading
0.865
Theta
0.789
Eigenvalue
3.088
Percentage Explained Variation
44.108
Single Factor Solution
Low Ranking = More Vulnerable
High Ranking = More Resilient
The principal components analysis produced a single factor solution with an eigenvalue of 3.088
and an explained variation of 44.108 percent. The Armor’s Theta Reliability coefficient for this
index was 0.789 reflecting adequate levels of internal consistency. The factor loadings for the
analysis ranged from -0.486 to 0.865. The strongest loadings in the analysis were percentage
change in median rent (0.865), percentage change in home values (0.792), and percentage moved
52
into house from 1999 to 2000 (0.728). All the factor loadings were above 0.350 and so were
included in the index.
Variables 1 and 2 were derived from the 1990 and 2000 U.S. Decennial Census Summary Tape
File 3. Variable 3 was downloaded from “City-Data.com” and the 2000 U.S. Decennial Census
Summary Tape File 3. Variables 4 and 5 are from the 2000 U.S. Decennial Census Summary
Tape File 3. Variables 1 and 2 were calculated by taking the 2000 value for median rent and
mortgage and subtracting the 1990 values, and then dividing the result by the 1990 values and
multiplying by 100. Variable 3 was calculated by taking the 2007 home value and subtracting
the 2000 value and then dividing the result by the 2000 value and multiplying by 100. Variable
4 was calculated by taking the number of renters in 2000 and subtracting the number of renters in
1990 and then dividing the result by the number of renters in 1990 and multiplying by 100.
(J) Personal Disruption Index
The personal disruption index (Table 23) consists of five variables: 1) percentage change in
unemployment 1990 to 2000 (range -0.31 to 1.78), 2) percentage change in travel time to work
1990 to 2000 (range -0.31 to 1.22 percent), 3) number of registered sex offenders per 1,000
population (range 0.02 to 6.87 per 1,000 population), 4) percentage of population separated
(range 1.4 to 4.4 percent), and 5) percentage of population that is divorced (range 8.5 to 17.7
percent).
The principal components analysis produced a single factor solution with an eigenvalue of 2.44
and an explained variation of 48.873 percent. The Armor’s Theta Reliability coefficient for this
index was 0.739 reflecting adequate levels of internal consistency. The factor loadings for this
analysis ranged from -0.382 to 0.928. The strongest loading in the analysis were for number of
registered sex offenders per 1,000 population (0.928), percentage population separated (0.871),
and percentage change in unemployment 1990-2000 (0.715). All the factor loadings were above
.350 and so were included in the index.
Variables 1 and 2 were derived from the 1990 and 2000 U.S. Decennial Census Summary Tape
File 3. Variable 3 was downloaded from “City-Data.com.” Variables 4 and 5 are from the 2000
U.S. Decennial Census Summary Tape File 3. Variables 1 and 2 were calculated by taking the
2000 unemployment and commuting time and subtracting the 1990 unemployment and
commuting time and dividing the result by the 1990 unemployment and commuting time and
multiplied by 100. Variable 3 was calculated by taking the number of sex offenders in a
community and dividing by the total population and then multiplying the result by 1,000.
53
Table 23: The Personal Disruption Index.
Community
Port Lavaca
Sea Drift
Port O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Little River
Mount Pleasant
Percentage
change in
unemployment
1990-2000
Percentage
change
travel time
to work
1990-2000
Number of
registered sex
offenders per
thousand pop
Percentage
separated
population
Percentage
divorced
population
Personal
Disruptions
Score
Ranking
-0.31
0.38
0.67
0.50
-0.25
1.78
0.80
-0.28
0.00
-0.08
0.47
0.22
1.22
0.70
0.27
0.53
0.01
0.19
0.15
-0.31
0.24
0.04
3.24
2.22
2.78
2.52
1.27
6.87
2.67
3.16
4.45
3.56
0.02
3.0
2.7
1.7
2.6
2.0
4.4
3.6
3.2
3.0
1.9
1.4
11.2
13.4
17.7
8.5
15.7
16.1
12.6
13.1
15.1
14.3
9.2
-0.220
-0.468
-0.172
-0.309
-0.849
2.406
0.508
-0.184
0.591
-0.304
-1.182
6
9
4
7
10
1
3
5
2
7
11
-0.381
0.928
0.871
0.409
PC components
Factor Loading
0.715
Theta Reliability
0.739
Eigenvalue
2.444
Percentage Explained Variation
48.873
Single Factor Solution
(K) Commercial Fishing Reliance Index
High Ranking = More Vulnerable
Low Ranking = More Resilient
The commercial fishing reliance index (Table 24) consists of five variables: 1) percentage of
labor force employed in agriculture, fishing, and hunting in 2000 (range 1.24 to 20 percent), 2)
pounds of landings per 1,000 population in 2007 (range 7,727 to 2,428,921 pounds per 1,000
population), 3) commercial fishing permits per 1,000 population in 2007 (range 0.01 to 15.38 per
1,000 population), 4) value of landings per 1,000 population (range $11,222 to $4,845,564), and
5) number of permitted dealers with landings per 1,000 persons (range 0.11 to 4.86).
The principal components analysis produced a single factor solution with an eigenvalue of 3.512
with an explained variation of 70.202 percent. The Armor’s Theta Reliability coefficient was
0.894 reflecting a high level of internal consistency. The factor loadings ranged from 0.765 to
0.902. The strongest factor loadings in the analysis were for the variables percentage of labor
force employed in agriculture, fishing, and hunting in 2000 (0.902), number of permitted dealers
with landings per 1,000 persons (0.873), commercial fishing permits per 1,000 population in
2007 (0.871). All the factor loadings were above 0.350 and so were included in the index.
Variable 1 was from the 2000 U.S. Decennial Census Summary Tape File 3. Variables 2
through 5 were from a custom data set generated by NOAA NMFS for this research. Variables 2
54
through 5 were standardized by taking the absolute occurrence for each variable in a community
and dividing by the total population and multiplying the result by 1,000.
Table 24: The Commercial Fishing Reliance Index.
Community
Percentage
employed in
Ag, Fishing, &
Hunting
Pounds of
Landings per
1,000
persons
Commercial
Fishing
Permits per
thousand pop
Value of
Landings
Per 1,000
population
Dealers with
Landings
per 1,000
persons
Commercial
Reliance
Index
Ranking
Port Lavaca
Sea Drift
Port O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Little River
Mount Pleasant
3.61
17.70
20.00
10.05
2.45
5.12
1.53
1.24
2.37
4.18
1.48
9,555
941,393
124,196
2,428,921
243,443
749,085
102,372
16,933
14,632
58,246
7,727
3.23
1.40
15.38
11.88
0.35
0.20
1.05
0.02
0.01
2.03
0.12
$34,638
$1,424,031
$350,978
$4,845,564
$211,624
$1,835,321
$222,314
$42,757
$11,222
$87,552
$15,504
0.26
2.1
4.86
2.38
0.44
1.82
0.25
0.11
0.13
0.79
0.2
-0.498
0.624
1.575
2.057
-0.543
0.149
-0.644
-0.76
-0.716
-0.470
-0.744
6
3
2
1
7
4
8
10
9
5
11
PC components
Factor Loading
0.902
Theta Reliability
0.894
Eigenvalue
3.512
Percentage Explained Variation
70.202
0.765
0.871
0.77
Single Factor Solution
(L) Recreational Fishing Reliance Index
0.873
High Ranking = More Reliant
Low Ranking = Less Reliant
The recreational fishing reliance index (Table 25) consists of five variables: 1) charter permits
per 1,000 population in 2007 (range 0 to 26.72 per 1,000 population), 2) marinas and related
businesses per 1,000 population (range 0 to 2.43), 3) marinas and related businesses jobs per
1,000 population (0 to 8.1 per 1,000 population), 4) marinas and related businesses gross
earnings per 1,000 population (range 0 to $449,392), and 5) boat launches per 1,000 population
(range 0.09 to 4.05 per 1,000 population).
The principal components analysis produced a single factor solution with an eigenvalue of 4.128
and an explained variation of 82.551 percent. The Armor’s Theta Reliability coefficient for this
index is 0.947 reflecting very high levels of internal consistency. The factor loadings for this
index range from 0.816 to 0.966. The strongest factor loadings in the analysis were for the
variables marinas and related businesses per 1,000 population (0.966), marinas and related
businesses jobs per 1,000 population (0.940), and charter permits per 1,000 population in 2007
(0.933). All the factor loadings were above 0.350 and so were included in the index.
55
Variables 1 and 5 are from a NOAA NMFS custom database. Variables 2, 3, and 4 are from the
2002 Economic Census available from the “American FactFinder” website. All the variables
were standardized by taking the absolute occurrence for each variable in a community and
dividing by the total population and multiplying the result by 1,000.
Table 25: The Recreational Fishing Reliance Index.
Community
Charter
Permits
per
thousand
pop
Marinas and
Related
Businesses
per 1,000
population
Marinas and
Related
Businesses
Jobs per
thousand pop
Marinas and
Related
Businesses
Gross $ per
thousand pop
Boat
Launches
per
thousand
pop
Recreational
Reliance
Index
Ranking
Port Lavaca
Sea Drift
Port O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Little River
Mount Pleasant
0.52
0.00
26.72
0.00
0.00
0.81
0.74
0.00
0.00
2.26
0.37
0.44
0.00
2.43
0.40
1.15
0.81
0.19
0.20
0.00
0.56
0.25
2.01
0.00
8.10
0.99
3.92
3.64
1.62
0.79
0.00
4.74
1.96
$71,921
$0
$449,392
$59,382
$182,879
$384,304
$131,717
$31,520
$0
$248,194
$115,907
0.44
2.10
4.05
0.20
0.26
0.81
0.21
0.09
0.25
0.34
0.09
-0.355
-0.517
2.763
-0.516
0.188
0.449
-0.401
-0.656
-0.816
0.244
-0.401
5
9
1
8
4
2
6
10
11
3
7
0.966
0.940
PC components
Factor Loading
0.933
Theta Reliability
0.947
Eigenvalue
4.128
Percentage Explained Variation
82.551
0.880
Single Factor Solution
(M) Social Fishing Dependence Index
0.816
High Ranking = More Reliant
Low Ranking = Less Reliant
The social fishing dependence index (Table 26) consists of five variables: 1) percentage of water
cover in the municipal boundary (range 0 to 77.8 percent), 2) boat launches per 1,000 population
(range 0.09 to 4.05 per 1,000 population), 3) percentage of labor force employed in agriculture,
fishing, and hunting in 2000 (range 1.24 to 20 percent), 4) marinas and related businesses per
1,000 population (range 0 to 2.43), and 5) number of permitted dealers with landings per 1,000
persons (range 0.11 to 4.86).
The principal components analysis produced a single factor solution with an eigenvalue of 3.438
and an explained variation of 68.780 percent. The Armor’s Theta Reliability coefficient was
0.886 for this index reflecting high levels of internal consistency. The factor loadings for this
index ranged from 0.355 to 0.982. The strongest factor loadings were for the variables number
of permitted dealers with landings per 1,000 persons (0.982), boat launches per 1,000 population
56
(0.942), and percentage of labor force employed in agriculture, fishing, and hunting in 2000
(0.938). All the factor loadings were above 0.350 and so were included in the index.
Variable 1 and 3 are from the 2000 Decennial Census Summary Tape File 3. Variables 2, 4, and
5 are from a custom NOAA NMFS database. Variable 1 was derived from Census data by
dividing total land area in the municipality into the percentage water cover. Variables 2, 4 and 5
were standardized by taking the absolute occurrence for each variable in a community and
dividing by the total population and multiplying the result by 1,000.
Table 26: The Social Fishing Dependence Index.
Community
Port Lavaca
Seadrift
Port O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Little River
Mount Pleasant
% Watercover
28.70
0.00
25.00
3.70
73.40
5.80
77.80
62.80
0.00
3.70
15.35
PC components
Factor Loading
0.355
Theta Reliability
0.886
Eigenvalue
3.438
Percentage Explained Variation
68.780
Boat
Launches
per
thousand
pop
0.44
2.10
4.05
0.20
0.26
0.81
0.21
0.09
0.25
0.34
0.09
0.942
% Employed
in Ag,
Fishing, &
Hunting
3.61
17.70
20.00
10.05
2.45
5.12
1.53
1.24
2.37
4.18
1.48
0.938
Marinas and
Related
Businesses
per 1,000
population
0.44
0.00
2.43
0.40
1.15
0.81
0.19
0.20
0.00
0.56
0.25
0.778
Single Factor Solution
Dealers
with
Landings
per 1,000
persons
0.26
2.1
4.86
2.38
0.44
1.82
0.25
0.11
0.13
0.79
0.2
Social
Dependence
Index
Ranking
-0.420
0.695
2.623
0.381
-0.374
0.197
-0.805
-0.801
-0.586
0.244
-0.401
7
2
1
3
6
4
11
10
8
5
9
0.982
High Ranking = More Reliant
Low Ranking = Less Reliant
(N) Local Fish Stock Sustainability Index (L-FSSI) and Quotient Indicators
The FSSI was developed in 2005, although NMFS has been collecting and reporting data on
fishery sustainability for about a decade (Buck, 2007). Collected at the national and regional
level (the eight fisheries management regions in the U S.), the FSSI is one of several assessment
and accountability tools that the NMFS uses to measure the effectiveness and efficiency of the
agency. The FSSI along with other measures are reported to the U.S. Office of Management and
Budget (OMB) that in turn gives the agency an overall efficiency rating based on setting
ambitious goals, achieving results, and managerial effectiveness. The FSSI is a performance
57
measure that assesses the sustainability of 230 fish stocks that are important to commercial and
recreational sectors. Most of the fish species are measured as individual stocks however a few
are measured as a stock complex (NMFS, 2008). Each stock is given a score of up to four if the
stock is sustainable. Points are awarded for the following specific criteria: (1) “overfished”
status is known = 0.5, (2) “overfishing” status is known = 0.5, (3) overfishing is not occurring =
1, (4) stock biomass is above “overfished” level defined for the stock = 1, and (5) stock biomass
is at or above 80% of the biomass that produces maximum sustainable yield (Bmsy)2 = 1 (NMFS,
2008). Since there are 230 stocks that are assessed by the FSSI the best score that can be
achieved is 920 (230x4). For the first quarter of 2008 the FSSI score was 531 meaning that the
agency could improve fisheries by a total of 389 points (NMFS, 2008). By this method NMFS is
able to show improvements or declines in their performance of the management the nation’s
fisheries.
Table 27: The Local Fish Stock Sustainability Index (L-FSSI) and Quotient Indicators.
Community
Weight
L-FSSI
Ranking*
Port Lavaca
3.09
4
Seadrift
2.88
2
Port O'Connor
3.21
5
Palacios
3.97
11
Seabrook
3.67
10
San Leon
3.39
7
Galveston
3.55
8
Texas City
3.21
6
Bacliff
3.00
3
Little River
1.07
1
Mount Pleasant
3.63
9
* 1 less sustainable, 11 = most sustainable
Community
Port Lavaca
Seadrift
Port O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Little River
Mount Pleasant
Second by
Landings
Species
Bait Shrimp
Eastern Oyster
Bait Shrimp
White Shrimp
White Shrimp
White Shrimp
Red Snapper
Bait Shrimp
Vermilion Snapper
Swordfish
Value
L-FSSI
Ranking
3.15
2.85
2.89
3.98
3.39
3.13
3.42
3.17
3.00
1.23
3.64
6
2
3
11
8
5
9
7
4
1
10
Weight
Quotient
Value
Quotient
56.65
2.29
1156.00
3.07
7.47
4.18
46.11
1203.00
17.10
8.03
190.60
1.38
257.00
1.07
0.96
0.44
11.11
1835.00
10.06
5.32
58
First by
Landings
Species
Eastern Oyster
Blue Crabs
Eastern Oyster
Brown Shrimp
Eastern Oyster
Eastern Oyster
White Shrimp
Eastern Oyster
Blue Crabs
Gag Grouper
White Shrimp
Third by
Landings
Species
Croaker
Black Drum
Brown Shrimp
Brown Shrimp
Scamp
Brown Shrimp
Weight
Quotient
Value
Quotient
55.64
21.53
29.52
9.63
20.84
37.91
8.30
33.64
24.39
20.88
2.75
7.99
11.45
5.68
2.95
7.02
8.43
2.41
7.33
15.22
12.96
1.94
Weight
Quotient
Value
Quotient
56.27
19.88
1.53
2.29
28.43
4.91
18.76
14.14
0.19
0.48
17.59
2.66
(1) Creation of the L-FSSI and Examples of Interpretation
The development of the L-FSSI is dependent upon assembling local community landings data.
In general, using landings to predict future impacts can be risky as you are looking back at what
was, rather than examining the fishery’s current state. Additionally, it is possible for landings to
be relatively stable even as overfishing is occurring and the fishery becomes less sustainable.
However, adding information from the FSSI to local landings data can help strengthen landings
as a better outcome and predictive variable for fisheries management.
An important issue in using community-level landings data revolves around federal
confidentiality rules. NMFS does not allow reporting landings data when there are less than
three fishers, processors, or distributors in a given community (Impact Assessment, 2005b). The
“rule of three” protects confidentiality by prohibiting the reporting of information that might be
attributed to a single business or individual. This keeps potential competitors from gaining
inside information about the activities of that business or individual (Impact Assessment, 2005b).
There are many small rural communities that have only one or two fish processors that contribute
a relatively large amount of jobs and income to the local economy (Impact Assessment, 2005b).
Nonetheless the data cannot be reported because of the rule of three. In many cases this
essentially makes community-level landings data unavailable to researchers outside of NMFS
because of the sensitive and confidential nature of the information. However, since the L-FSSI
is an aggregate of all landings data and does not report information by individual species, it
would not violate federal confidentiality rules.
Although the L-FSSI score for an individual community would not violate federal confidentiality
rules, we need to report community-level landings data by species to show exactly how the
measure is constructed. As such we have decided to present fictitious data that represents an
amalgam of several Gulf coast fishing communities. This will allow us to illustrate the logic of
the FSSI index without violating federal confidentiality rules. Shrimpville (fictitious place and
data) represents a typical shrimping community on the Gulf of Mexico region (Table 28). The
majority of landings come from brown (Farfantepenaues aztecus) and white shrimp
(Litopenaeus setiferus). There are some minor landings of finfish. If we were to take all of the
FSSI scores for all the species fished in Shrimpville and summed them, the total comes to 17.
Since there are five different species that were landed the average FSSI = 3.40 (Sum of FSSI 17 /
Number of Species 5 = 3.40). However, this misrepresents the real FSSI of the community since
red snapper (Lutjanus campechanus), which only makes up 0.21% of landings has an FSSI of 1.
By taking the pounds reported of landings for each species and dividing them by the sum of all
landings, a percent of total landings by species is computed. For example, brown shrimp had
1,991,578 lbs reported and is divided by 3,205,703 lbs total landings to represent 62.13% of all
landings in Shrimpville. When the percent of landings is multiplied by the FSSI score for each
species the sum of that product is the L-FSSI. This L-FSSI is a weighted average FSSI that
considers all species in the local fishery. It returns to the original scoring metric, which ranges
from 0 to 4. The L-FSSI in Shrimpville is 3.99, which indicates the fishery is very sustainable.
59
By comparison Fintown (which is a fin-fishing reliant community) is far more vulnerable. The
L-FSSI in Fintown is 1.62 indicating that the species that the fishers in the community rely upon
are less sustainable. This suggests that the community is vulnerable to any changes in the
fishery.
Table 28: Theoretical community landings data with Gulf of Mexico landings data.
Shrimpville (Fictitious Place and Data)
Species
Shrimp, Northern Brown
Shrimp, Northern White
Snapper, Red
Cobia
Snapper, Vermilion
Sum
6773
37
3,205,703
3.4
Sum
89,076
332,145
0.0004
0.03
0.37
41.95
1.49
0.0000
L-FSSI
0.57
FSSI
1
17.84
0.1784
1.49
11.97
1.5
16.48
0.2473
1.53
10.77
6.51
11.42
12.94
3
10.52
0.5
24.29
12.5
100.00
L-FSSI
0.0651
0.1713
0.5174
0.3156
0.1214
0.29
1.40
0.55
0.34
0.47
1.62
% pounds
% pounds*FSSI
4
139,955,385
44.79
1.7915
Grouper, Red
4
6,081,226
1.95
0.0778
Shrimp, Pink
Crab, Florida Stone Claws
Snapper, Red
Lobster, Caribbean Spiny
Snapper, Vermilion
Mackerel, Spanish
4
1.5
1
1.5
4
4
0.00
3.99
Shrimp, Brown
4
0.14
L-FSSI
Quotient
Average FSSI 1.56
Gulf of Mexico Data 2006 (Real Data)
Species
FSSI
Pounds
Shrimp, White
0.90
Gulf
(region)
% pounds
4
1,367,643
Quotient
1.39
Fintown
(local)
% pounds*FSSI
1.5
176,921
2.4850
% pounds
44.79
Fintown (local)
% pounds
1
156,209
143,877
100
L-FSSI
0.0021
0.00
17
Gulf
(region)
1.5060
0.21
4
225,437
Gag
0.01
37.65
1
Crab, Florida Stone Claws
Mackerel, King
4
4
243,978
Mackerel, Spanish
62.13
360
1,206,955
Snapper, Red
Lobster, Caribbean Spiny
4
1,991,578
Pounds
Reported
Amberjack, Greater
Shrimpville
(local)
% pounds*FSSI
FSSI
Average FSSI
Fintown (Fictitious Place and Data)
Species
Shrimpville
(local)
% pounds
Pounds
Reported
131,095,568
41.95
12,485,948
4.00
4,789,141
1.53
4,645,016
1.49
4,367,510
1.40
1,769,801
1,732,888
0.57
60
0.55
1.6781
0.1598
0.0230
0.0149
0.0210
0.0227
0.0222
22.46
8.16
23.52
30.94
51.67
Gag
Snapper, Yellowtail
1,458,224
0.47
0.0023
1
908,189
0.29
0.0029
4
Mackerel, King
3
Amberjack, Greater
Tuna, Little Tunny
1.5
Shrimp, Royal Red
1.5
Dolphinfish
4
Cobia
4
Triggerfish, Gray
0.5
Drum, Red
Average FSSI
0.5
1.5
2.67
2. Ground-Truthing
1,154,007
0.37
1,058,990
0.34
319,573
0.10
293,981
0.09
225,073
0.07
93,609
0.03
32,778
0.01
22,192
312,489,099
0.01
100.00
Weighted Average
0.0148
0.0102
0.0015
0.0014
0.0029
0.0012
0.0001
0.0001
3.85
(A) Interviews and Coding
The development of the codes and their analyses are presented in the sections below.
(1) Master Sheet and Codes
A master-sheet was created to include the total number of references to each topic across all of
the interviews by community (abbreviated for tabular format). A copy of the master-sheet is
provided as Table 29. The table is divided into two sections, FISHERIES and INDIVIDUALS,
determined by the focus of the interview question and derived code. Each of the sections is
further divided according to the major topics that collectively constitute the section. The
FISHERIES Section includes nine sub-sections: Infrastructure, Catch Levels, Income/Pricing,
Operating Expenses, Regulations, Institutions, Gear and Boat Changes, Land/Water Use
Changes, and Place of Community. The INDIVIDUALS Section includes nine sub-sections:
Fishing Experience, Multiple Fisheries, Ownership, Sources of Income, Changes in Liquidity,
Skills/Education, Job Satisfaction, and Entrepreneurship. In almost all instances, the subsections are further divided into categories, and in some instances, those are sub-divided yet
again (Decline in labor availability, e. g., contains four sub-categories).
The totals for each sub-section are highlighted in yellow at the top of the sub-section, and the
totals for each section are highlighted in green, again at the top of the section. The totals for the
two sections, FISHERIES and INDIVIDUALS are highlighted at the bottom of the table in blue,
as in all instances, with scores for each of the five communities.
Chi square tests were run independently on the FISHERIES and the INDIVIDUALS sections of
Table 29, using matrices constructed from the communities by the sub-sections (the totals in
yellow). In the case of FISHERIES, χ2 = 103.82, and with 32 degrees of freedom, p < 0.001.
The results for INDIVIDUALS were χ2 = 98.10, and with 32 degrees of freedom, p < 0.001.
61
Table 29: References by keywords to code topics by individuals and community.
GalvBay PLavaca Seadrift Palacios O'Connor Totals
160
126
225
214
158
883
FISHERIES
Infrastructure
45
43
61
60
60
269
Decline # fishers
28
13
17
16
15
89
Decline # boats
5
13
16
15
13
62
Decline # employees
0
4
1
10
6
21
Decline # processors/buyers
0
3
6
6
8
23
Decline # public docks
0
3
3
1
7
14
Decline # bait shops
0
0
0
0
2
2
Decline in labor availability
0
0
3
0
1
4
recruitment difficulties
1
2
7
6
5
21
aging population
10
2
0
3
3
18
no locals for crew
1
2
7
2
0
12
inexperienced workers
0
1
1
1
0
3
Catch Levels
15
18
32
17
18
100
Lower volume
10
9
15
12
12
58
Lower CPUE
5
9
17
5
6
42
Income/Pricing
15
16
30
23
14
98
Import price lowering
15
3
11
16
6
51
Decline in ex vessel price
0
7
7
7
5
26
Monopolistic pricing
0
6
12
0
3
21
Operating Expenses
20
8
19
35
13
95
High fuel prices
20
6
16
16
8
66
Effort to buy fuel in
Mexico
0
1
0
7
1
9
No credit/cash only
0
1
2
2
2
7
Increases for cost of crew
0
0
0
3
0
3
Boat maintenance neglect
0
0
1
5
2
8
Other
0
0
0
2
0
2
Regulations
34
4
10
8
5
61
Unspecified
23
0
5
0
0
28
TEDs/Fish-eyes
4
1
0
5
1
11
Limited entry
5
0
2
2
2
11
Buy back
1
0
0
1
2
4
Homeland Security
1
1
2
0
0
4
EEZ
0
2
1
0
0
3
Institutions
5
13
20
22
0
60
Foreclosures
0
3
1
7
0
11
Loans
0
2
4
3
0
9
Banks
0
2
1
3
0
6
Bankruptcy
0
3
0
3
0
6
Local governments
0
1
3
1
0
5
62
Gear & Boat Changes
Barge construction
Yacht retrofitting
Increase in net # and size
Other
Land/Water Use Changes
Increase in pollution
Increase in taxes
Increase in condos/hotels
Increase in tourism
Expansion of industry
Place in Community
No longer fishing comm’ty
Fishing present but reduced
Less fishing, neg impact
0
0
0
0
0
7
7
0
0
0
0
19
11
8
0
3
0
0
3
0
21
10
1
1
2
7
0
0
0
0
1
0
0
1
0
29
12
3
1
6
7
23
17
0
6
9
3
4
1
1
23
2
7
10
4
0
17
2
12
3
3
0
0
3
0
31
4
12
8
6
1
14
5
9
0
16
3
4
8
1
111
35
23
20
18
15
73
35
29
9
INDIVIDUALS
Fishing Experience
Multi-generational
Family network
First generation
Multiple Fisheries
Unspecified
Oyster
Bait
Other
Ownership
Boat
Quota
Processor
Ice house
Sources of Income
Other: non-fishing
Fishing only
Family/wife/children
No Social security
Changes in Liquidity
All assets in fishing
Ability to sell boats
Debt Levels/Defaults
Skills/Education
Marketable skills
Education level
139
30
22
8
0
3
0
3
0
0
20
9
5
5
1
39
12
23
4
0
5
5
0
4
19
19
0
94
14
7
7
0
21
6
6
4
5
4
4
0
0
0
18
7
4
5
2
4
1
3
2
23
9
9
122
28
16
11
1
21
8
2
8
3
13
8
3
0
2
20
6
8
6
0
7
5
2
0
20
7
10
81
14
7
7
0
6
0
4
1
1
12
7
0
3
2
20
14
3
2
1
2
1
1
2
12
8
3
61
8
2
0
6
9
4
4
1
0
5
0
4
1
0
15
12
1
0
2
4
1
3
0
6
5
1
497
94
54
33
7
60
18
19
14
9
54
28
12
9
5
112
51
39
17
5
22
13
9
8
80
48
23
63
Language skills
Job Satisfaction
Entrepreneurship
TOTAL
0
12
7
299
5
5
3
220
3
12
1
347
1
2
11
295
0
5
9
219
9
36
31
1380
(2) Explanations of Codes
Although many of the labels/terms in Table 29 are self-explanatory, e.g., Decline # of fishers,
some will require additional description and explanation. Those are identified and described
below, in the order in which they appear in Table 29.
•
•
•
•
•
•
•
•
•
•
•
•
Low import price refers to the deflation of ex-vessel price of shrimp due to the volume
and lower prices of imported fish.
Decline in ex-vessel price potentially overlaps with Low import price, but landing values
for some species have declined independently.
Other refers to any Operating Expenses other than the ones listed. Examples are the
increased costs incurred from having to stay on the water longer to make profitable levels
of catch and the increased costs of boat repairs.
Monopolistic pricing refers to instances in which respondents noted that they have
problems selling their catch at market rates, given that they have access to only one
dealer or buyer, who can set the buying price at any chosen, i.e., monopolistic level.
Homeland Security refers mainly to problems of recruiting workers but also to difficulties
encountered on re-entering US waters after trips to Mexico to buy cheaper fuel.
EEZ refers to the additional constraints/regulations placed on fishers when they enter
federal waters beyond three miles offshore.
TPWD refers to constraints/regulations imposed by the Texas Parks and Wildlife
Division (typically seen negatively).
Banks refers to the unwillingness of banking institutions to provide loans for fishers,
especially in times of crisis and thus critical needs.
Barge construction and yacht retrofitting refer to observed trends in new boat
construction businesses at docks and marinas, in which barges are constructed and fishing
boats are retrofitted into pleasure yachts, reflecting a de-emphasis in commercial fishing
toward other commercial or recreational interests.
Place in Community refers to whether the respondents view the community now as a
fishing community or whether it has changed to the point that the term no longer applies.
Multiple Fisheries refers to instances in which an individual is currently or in the past has
participated in more than one fishery.
Changes in Liquidity refers to whether the fisher has assets that can be sold or converted
into cash.
64
•
•
Skills/Education was seen as factors that could enable a fisher to pursue other livelihood
options, either as career or as providing access to other income. Conversely, their
absence meant that fishers had no options for other livelihoods or livelihood support.
Entrepreneurship is a cover term that includes reports of instances in which fishers saw
and took opportunities to improve their incomes and livelihood status.
(3) Coding Results – Raw Scores
To focus initially only on major similarities and differences across the communities, a striking
result is the overall similarities of concerns among the fishers in the five communities. In all five
of them, Infrastructure received the most comments, by considerable margins. Also within
FISHERIES declines in Catch Levels, Income/Pricing, and Land/Water Use Changes all received
approximately 100 comments, and Operating Expenses, Place in Community, Institutions and
Regulations received comments in the range of 60-75. Only Gear and Boat Changes showed
low scores across all of the communities.
The section INDIVIDUALS also shows consistent levels of concerns across the sub-sections and
categories. Sources of Income and Fishing Experience each scored approximately at 100,
whereas Skills/Education was at 80 and had somewhat similar scores across the communities.
Multiple Fisheries and Ownership were in the range of 54-60, and Changes in Liquidity, Job
Satisfaction, and Entrepreneurship were in the range of 22-36. Only Debt Levels/Defaults had
consistently low scores across communities and a total of only 10.
Among the more obvious differences across communities is the large number of zeros, or no
comments, by the fishers in the Galveston Bay communities. Twenty-nine of the topics in
FISHERIES had zero comments, as opposed 19 zeros for Port O’Connor, 15 for Port Lavaca, 10
for Seadrift, and only 7 for Palacios. This, again, is a function of the difficulty in obtaining
detailed interviews in the Galveston Bay communities. Interestingly, however, the Galveston
Bay communities had zero comments on the topics of Decline # employees, Decline #
processors/buyers, and Decline in # public docks, unlike all of the other communities. Another
possible contributing factor to those differences is that as an urbanized area, the Galveston Bay
communities had already undergone those infrastructural changes, i.e., earlier than the
communities in the Matagorda and San Antonio Bays, and that those infrastructural features had
stabilized. Possible supportive factors are that Galveston Bay had a higher score on aging
population, and the score for Decline # fishers was higher than the other communities.
Similarly, the zeros in Land/Water Use Changes were much higher in all of the other
communities, which again could be exacerbated by interviewing difficulties but could also be
due in part to the fact that those features had already changed and stabilized to greater extents in
the more urban Galveston Bay region.
65
The other category in which Galveston Bay fishers did not comment was Operating Expenses,
except for High fuel prices, possibly due to operating expenses having been higher in the urban
area for longer periods of time.
Two additional differences in scores in FISHERIES were salient and tended to differentiate
Galveston Bay from the other communities. Within Income/Pricing, Galveston Bay had a much
higher score on Import price lowering in comparison with Port Lavaca, Seadrift, Palacios, and
Port O’Connor, 115 as opposed to 18, 55, 23, and 30, respectively. Fishers in Galveston Bay
were 2 to 6.5 times more likely to comment on the lowering of the prices of fish due to imports.
Conversely, Galveston Bay fishers also scored lower on Land/Water Use Changes, again likely
due to the more urban environment.
The section INDIVIDUALS showed fewer differences across the communities as compared with
FISHERIES. Sources of Income was almost twice as high in Galveston Bay as in the other
communities. In addition, Galveston Bay fishers reported that their income was from Fishing
only at a higher rate, 23 as opposed to 4 in Port Lavaca, 8 in Seadrift, 3 in Palacios, and 1 in Port
O’Connor. Those scores suggest that the fishers in Galveston Bay are more likely to be full-time
fishers than is the case in the other bays. The scores for Job Satisfaction were also considerably
higher for Galveston than for the other bays, lending support to the idea that Galveston Bay
fishers were specialists constrained to a greater extent on fishing, as opposed to fishers in the
other communities. Supportive evidence may come from the distribution of scores in the
Skills/Education section, where Galveston Bay fishers made no comments on Education level
and Language skills, in contrast to the other communities.
(4) Coding Results – Rate Measures
The differences noted above are all based on the raw scores of the coding totals, extracted from
the interviews. A different set of scores can be presented, however, by dividing the totals in each
cell by the number of individuals interviewed, thereby providing a rate measure. Those scores
are directly comparable, as they represent the number of coded items per individual. The results
are given in Table 30, multiplied by 100 to avoid decimal fractions.
Table 30: Rate measures of topic references by individuals and community.
GalvBay Lavaca Seadrift Palacios O'Connor
450
745
996
1646
790
FISHERIES
Infrastructure
130
254
233
461
300
Decline # fishers
78
76
85
123
75
Decline # boats
14
76
8
115
65
Decline # employees
0
24
5
77
30
Decline # processors/buyers
0
18
30
46
40
Decline # public docks
0
18
15
8
35
Decline # bait shops
0
0
0
0
10
Decline in labor availability
0
0
15
0
5
66
Totals
4627
1378
437
350
136
134
76
10
20
recruitment difficulties
aging population
no locals for crew
inexperienced workers
Catch Levels
Lower volume
Lower CPUE
Income/Pricing
Import price lowering
Decline in ex vessel price
Monopolistic pricing
Operating Expenses
High fuel prices
Effort to buy fuel in Mexico
No credit/cash only
Increases for cost of crew
Boat maintenance neglect
Other
Regulations
Unspecified
TEDs/Fish-eyes
Limited entry
Buy back
Homeland Security
EEZ
Institutions
TPWD
Foreclosures
Loans
Banks
Bankruptcy
Local governments
Gear & Boat Changes
Barge construction
Yacht retrofitting
Increase in net # and size
Other
Land/Water Use Changes
Increase in pollution
Increase in taxes
Increase in condos/hotels
Increase in tourism
Expansion of industry
3
28
3
0
42
28
14
42
42
0
0
56
56
0
0
0
0
0
94
64
11
14
3
3
0
14
14
0
0
0
0
0
0
0
0
0
0
19
19
0
0
0
0
12
12
12
6
106
53
53
94
18
41
35
47
35
6
6
0
0
0
24
0
6
0
0
6
12
78
12
18
12
12
18
6
18
0
0
18
0
124
59
6
6
12
41
67
35
0
35
5
160
75
85
150
55
35
60
23
8
0
10
0
5
0
65
40
0
10
0
10
5
100
55
5
20
5
0
15
5
0
0
5
0
145
60
15
5
30
35
46
23
15
8
131
92
39
177
23
54
0
269
123
54
15
23
39
15
62
0
39
15
8
0
0
169
38
54
23
23
23
8
70
23
31
8
8
177
15
54
77
31
0
25
15
0
0
90
6
3
70
30
25
15
65
40
5
10
0
10
0
25
0
5
10
10
0
0
0
0
0
0
0
0
0
15
0
0
15
0
155
20
60
40
30
5
121
78
70
19
529
308
221
533
268
55
110
460
334
65
41
23
54
15
270
104
61
49
21
19
17
361
119
77
55
40
41
29
108
23
31
46
8
620
173
135
128
103
81
Place in Community
No longer fishing comm’ty
Fishing present but reduced
Less fishing, neg impact
53
31
22
0
0
0
0
0
115
85
0
30
13
15
92
23
70
25
45
0
368
156
159
53
INDIVIDUALS
Fishing Experience
Multi-generational
Family network
First generation
Multiple Fisheries
Unspecified
Oyster
Bait
Other
Ownership
Boat
Quota
Processor
Ice house
Sources of income
Other: non-fishing
Fishing only
Family/wife/children
No Social security
Changes in Liquidity
All assets in fishing
Ability to sell boats
Debt levels/defaults
Skills/Education
Marketable skills
Education level
Language skills
Job Satisfaction
Entrepreneurship
TOTAL
386
83
61
22
0
8
0
8
0
0
56
25
14
14
3
108
33
64
11
0
14
14
0
11
53
53
0
0
33
19
836
547
82
41
41
0
123
35
35
24
29
24
24
0
0
0
100
41
24
29
6
24
6
18
12
135
53
53
29
29
18
1292
610
140
8
55
5
105
40
10
40
15
65
40
15
0
10
100
30
40
30
0
35
25
10
0
100
35
50
15
60
5
1606
405
70
35
35
0
30
0
20
5
5
60
35
0
15
10
100
70
15
10
5
10
5
5
10
60
40
15
5
10
55
2052
305
40
10
0
30
45
20
20
5
0
25
0
20
5
0
75
60
5
0
10
20
5
15
0
30
25
5
0
25
45
1095
2253
415
227
153
35
311
95
93
74
49
230
124
49
34
23
483
234
148
80
21
103
55
48
33
378
206
123
49
157
142
6881
Rate measures provide a more accurate measurement, given that the scores per community are
on the same basis, the number of topics/codes per interview. The community with the lowest
number of interviews (Palacios) will show a relatively higher score, whereas the community with
the largest number of interviews (Galveston Bay collectively) will show a relatively lower score.
68
To report the results from Table 30 initially in terms of total scores, Palacios had the highest
score (2052), Seadrift had the second highest (1606), and the other three communities were
considerably lower, Port Lavaca (1292), Port O’Connor (1095), and Galveston Bay (836).
The totals for FISHERIES are distributed across the communities in exactly the same order,
except that Port Lavaca and Port O’Connor are reversed: Palacios (1646), Seadrift (996), Port
O’Connor (790), Port Lavaca (745), and Galveston Bay (450).
The relative magnitudes,
however, differ, especially in terms of Palacios having a comparatively higher score in
FISHERIES than in the total of FISHERIES and INDIVIDUALS. The ratio between Seadrift and
Palacios in FISHERIES is 996/1646, or 61 percent, whereas the ratio for FISHERIES and
INDIVIDUALS combined is 1606/2052, or 78 percent. The point is that interviewees in Palacios
were comparatively more concerned and made more comments in FISHERIES by a larger
margin.
The codes within FISHERIES showing the greatest differences across communities are
Operating expenses, with Palacios having more than four times more comments as the next
closest community, Port O’Connor, 269 as opposed to 65, Institutions, with Palacios showing
169 as opposed to 100 for Seadrift, and Place in Community, where Seadrift has the most
comments, 115, as opposed to Port O’Connor with 70. Those results indicate that fishers in
Palacios primarily, and Seadrift secondarily, are having the greatest financial difficulties and that
interviewees in Seadrift and Port O’Connor have seen their communities change the most in
terms of no longer being fishing communities.
It is also noteworthy that the sub-sections in FISHERIES with the highest total scores were
Infrastructure (1450) and Land/Water Use Changes (620). The latter is a surprise, given that it
is higher than Income/Pricing (533), Regulations (532), and Catch Levels (529). The lowest
score was for Gear & Boat Changes (108), likely reflecting the financial constraints under which
fishers have to operate.
The total rate scores for INDIVIDUALS are distributed quite differently across communities,
compared with FISHERIES. The highest total is for Seadrift (610), followed by Port Lavaca
(547), Palacios (405), Galveston Bay (386), and Port O’Connor (305). The results suggest that
individuals in Seadrift are comparatively more concerned about the problems and issues facing
individual fishers than is the case in the other communities. Port O’Connor shows the least
amount of concern.
Within the section INDIVIDUALS, the code/topic showing the greatest differences across
communities is Multiple fisheries, with Port Lavaca having the highest score (123), in
comparison with the second-place community, Seadrift (105). The other three communities had
less than 50 each. Port Lavaca and Seadrift also have considerably higher scores for
Skills/Education, 135 and 100, respectively, than the other three communities, all 60 or below.
69
Those results suggest that interviewees in those two communities were comparatively more
sensitive to concerns about individuals.
The sub-sections that show the highest scores in the INDIVIDUALS section are Sources of
Income (483), with all communities scoring high, Fishing Experience (415), again with all
communities scoring high except for Port O’Connor, and Skills/Education (378). Those highlevel concerns appear to be related to ability or capacity to diversify within fisheries and within
the communities in terms of other income possibilities.
(5) Summary – Code Scores
The results of the totals for raw scores (total instances) and rate measures (instances per
interview) of codes show differences across the communities. Although the fishers in each
community were asked the same questions from the interview protocol, the numbers of responses
by topic (code) were not the same. The different totals of responses by topic can be taken as
indicators representing local concerns. Many of the concerns, but not all, are shared across the
five communities, but also not to the same degree. The major differences can be taken to
indicate the extent to which the topics are of special or, in some cases, unique concerns in the
respective communities.
Table 31: Comparison of rank of FISHERIES section totals by community.
TOPIC/KEYWORD
GalvBay
Lavaca
Seadrift
Palacios
O’Conner
Infrastructure
Catch Levels
Income/Pricing
Operating Expenses
Regulations
Institutions
Gear/Boat Changes
Land/Water Use
Place in Community
5
5
5
3
1
4
5
5
4
3
3
3
4
5
3
2
4
5
4
1
2
5
2
2
4
3
1
1
2
1
1
3
1
1
1
2
2
4
4
2
4
5
3
2
3
TOTALS
AVERAGE
37
4.11
32
3.56
24
2.67
13
1.45
29
3.22
Overall, Palacios shows the most concern with topics in FISHERIES, having the highest rank in
six of the nine sub-sections and an average score of 1.45. Seadrift is second, with the highest
score in two of the sub-sections, Catch Levels and Place in Community, and with an average
score of 2.67. The other three communities have considerably higher averages, and only
Galveston Bay has the highest rank for one sub-section, Regulations (Table 31).
Similar comparisons can be made for the section INDIVIDUALS, as presented in Table 32.
70
Table 32: Comparison of rank of INDIVIDUALS section totals by community.
TOPIC/KEYWORD
GalvBay
Lavaca
Seadrift
Palacios
O’Conner
Fishing Experience
Multiple Fisheries
Ownership
Sources of Income
Changes in Liquidity
Debt Levels/Defaults
Skills/Education
Job Satisfaction
Entrepreneurship
2
5
3
1
4
2
4
3
3
3
1
5
3
2
1
1
2
4
1
2
1
3
1
4.5
2
2
5
4
3
2
3
5
3
3
5
1
5
4
4
5
3
4.5
5
4
2
27
3.00
22
2.44
21.5
2.39
29
3.22
36.5
4.06
TOTALS
AVERAGE
Overall, Interviews from Seadrift fishers showed the most concern for INDIVIDUALS topics,
scoring first on three of the nine topics but having the lowest average, 2.39. Port Lavaca is a
close second, also with three top scores and an average of 2.44. Galveston Bay and Palacios are
third and fourth, respectively, and each scores first on one sub-section. Port O’Connor has the
highest average by a considerable margin, 4.06. Again, Port Lavaca and Seadrift are
communities in which fisheries have diminished saliently within the communities and in which
fishers have had to diversity their interests and livelihood strategies.
In summary, Palacios has been the site of major fisheries, especially bay and gulf shrimp, and the
fishers are under heavy duress to be able to continue to survive financially. The greatest concern
with FISHERIES is shown in that community. Seadrift has had a similar struggle, but fisheries
have not been as resilient there. Accordingly, Seadrift struggles with concerns about
FISHERIES and INDIVIDUALS, being first in the latter and second in the former. Port Lavaca
fisheries have been embedded in a community economy that is diversified and never as
dependent on fisheries as has been the case in the other communities. It is not surprising that
they are among the top two communities in concerns with topics in INDIVIDUALS.
(6) Keywords and Community Characteristics: Vulnerability and Resilience
The distribution of the keywords and the totals across topics reveals interesting information
about the characteristics of the five communities, as discussed. Further analysis is needed,
however, in relation to what the keywords indicate about vulnerability and resilience of the
fishers and communities. Vulnerability, again, is viewed as susceptibility to conditions that act
negatively on the resilience of individuals and thus sustainability of fishing communities.
Resilience is the ability to respond to external inputs that create perturbations in a system without
fundamental alteration of the system.
71
Each of the keywords in Table 29 was assessed as to whether they indicated vulnerability or
resilience for the fishers and their fishing communities. In most cases, the assessments were
straightforward and not difficult to make. Increase in pollution, for instance, can easily be seen
as vulnerability. Decline in # fishers, on the other hand, is not as clear, since a decline could be
seen as a reduction in fishing pressure and competition, allowing for better success among the
fishers remaining in a fishery. The responses in the interviews, however, were indicative of the
opposite, reflecting the view that the number of fishers had declined due to the varied difficulties
in making a living in the fishery. The loss in numbers of fishers was an indication of everyone’s
vulnerability to the recent and current conditions that work against sustainability. Almost all of
the keywords in the FISHERIES section, in fact, were indicators of vulnerability. That
information is given in Table 33, which contains the rate measures and totals for each
community.
Topics for which fewer than five fishers total made comments are omitted from Table 33, on the
grounds that only a small minority of the fishers considered the topic sufficiently important for
comment. Although some interesting developments, such as Barge construction in the marina at
Palacios, were thus omitted, those did not apply to other communities and were thus not included
in the table. Table 33 does show, however, which of the topics were of the broadest concern and
the differences in the rates of the concern across communities. As the table shows,
Infrastructure was the most widespread concern, followed, unexpectedly, by Land/Water Use
Changes. More predictable, Catch Levels was next, followed by Place in Community, Operating
Expenses, Institutions, and Regulations. The lower scores on the last two topics were also
somewhat unexpected.
In terms of community vulnerability, Palacios shows the highest total score, indicative of the
recent and current pressures that are reducing the sustainability of a once vibrant fishing
community. Three communities have intermediate-level scores, Seadrift, Port O’Connor, and
Port Lavaca, still placing them, however, comparatively high on vulnerability. Galveston Bay
has the lowest scores, but, again, that is likely to be a function of the interview difficulties.
Table 33: Topics/Keywords that indicate FISHERIES vulnerability.
TOPIC/KEYWORD
Infrastructure
Decline # fishers
Decline in labor
available
Decline # boats
Decline # employees
Decline # buyers
Decline # docks
GalvBay Lavaca Seadrift Palacios O’Conner TOTALS
130
78
34
254
76
42
233
85
90
461
123
92
290
75
45
1368
437
303
14
0
0
0
76
24
18
18
8
5
30
15
115
77
46
8
65
30
40
35
278
136
134
76
72
Land/Water Use
Changes
Increase in pollution
Increase in taxes
Increase in condos/hotels
Increase in tourism
Expansion of industry
19
124
146
177
155
621
19
0
0
0
0
59
6
6
12
41
60
16
5
30
35
15
54
77
31
0
20
60
40
30
5
173
136
128
103
81
Catch Levels
Lower volume
Lower CPUE
42
28
14
106
53
53
160
75
85
131
92
39
90
60
30
529
308
245
Place in Community
No longer fishing comm.
Fishing present/decline
Less fishing/negative
53
31
22
0
0
0
0
0
115
85
0
30
130
15
92
23
70
25
45
0
368
155
159
53
Operating Expenses
High fuel prices
Fuel from Mexico
Boat maintenance
20
20
0
0
41
35
6
0
13
8
0
5
216
123
54
39
55
40
5
10
345
226
65
54
Institutions
TPWD
Foreclosures/Bankruptcy
Banks/Loans
14
14
0
0
72
12
36
24
85
55
5
25
161
38
77
46
0
0
0
0
332
119
118
95
Regulations
Unspecified
TEDs/Fish-eyes
94
64
11
6
0
6
40
40
0
39
0
39
5
0
5
184
104
61
TOTALS
372
603
792
1315
665
3727
Only two of the keywords in the FISHERIES section indicate resilience, as indicated in Table 34.
Both of the topics, Limited entry and Buy back, are within the sub-section Regulations. The
total scores are also low for each topic, reflecting a low level of significance across the fisheries.
Unlike in Table 33, topics with fewer than five responses are included, given the low level on
instances.
73
Table 34: Topics/Keywords that indicate FISHERIES resilience.
TOPIC/KEYWORD GalvBay
Regulations
Limited entry
Buy back
TOTALS
14
3
17
Lavaca
Seadrift
Palacios
O’Conner
TOTALS
0
0
0
10
0
10
15
8
23
10
10
20
49
21
70
The INDIVIDUALS section was also assessed for vulnerability and resilience. As in
FISHERIES, some topics were clearer than others, and in the less clear cases, contextual
information from the interviews was taken into account. Fishing only, for example, could be an
indicator of resilience, if the fishing enterprise was sufficiently economically successful, but in
the context of the interviews, fishers tended to view it as a liability. On the other hand, Multiple
fisheries, could indicate vulnerability, but the fishers talked about participation in more than one
fishery, through time or at the same time, as positive ways to maintain their lives as fishers.
Accordingly, the keyword Multiple fisheries was taken to indicate resilience. The results for
vulnerability are given in Table 35 and the results for resilience are given in Table 36, below.
Relatively few of the codes/keywords indicated vulnerability, only four all together, and one of
those, First generation applied only to two communities (Port O’Connor and Seadrift) and with a
very small number of responses. No Social Security also had very few comments. The only
topic with substantial numbers was Fishing only in Galveston Bay, again consistent with
livelihood fishing specialization in those urbanized environments. Galveston Bay also scored
highest on All assets in fishing.
Table 35: Topics/Keywords that indicate INDIVIDUALS vulnerability.
TOPIC/KEYWORD
GalvBay
Lavaca
Seadrift
Sources of Income
Fishing only
No Social Security
64
64
0
30
24
6
40
40
0
20
15
5
15
5
10
169
148
21
Changes in Liquidity
All assets in fishing
14
6
25
5
5
55
Fishing Experience
First generation
0
0
5
0
30
35
TOTALS
78
36
70
25
50
259
74
Palacios O’Conner
TOTALS
Table 36: Topics/Keywords that indicate INDIVIDUALS resilience.
TOPIC/KEYWORD
GalvBay
Lavaca
Seadrift
Palacios
Skills/Education
Marketable skills
Education level
Language skills
53
53
0
0
135
53
53
29
100
35
50
15
60
40
15
5
30
25
5
0
378
206
123
49
Sources of Income
Other: non-fishing
Family/wife/children
44
33
11
70
41
29
60
30
30
80
70
10
60
60
0
314
234
80
Multiple Fisheries
8
123
105
30
45
311
Fishing Experience
Multi-generational
Family network
83
61
22
82
41
41
63
8
55
70
35
35
10
10
0
308
155
153
Ownership
Boat
Processor/Ice house
Quota
56
25
17
14
24
24
0
0
65
40
10
15
60
35
25
0
25
0
5
20
230
124
57
49
Job Satisfaction
33
29
60
10
25
157
Entrepreneurship
19
18
5
55
45
142
Changes in Liquidity
Ability to sell boats
0
18
10
5
15
48
296
499
468
370
255
1888
TOTALS
O’Conner TOTALS
In contrast to FISHERIES, where most of the topics can be seen as instances of vulnerability,
most of the topics in INDIVIDUALS can be seen as instances of resilience. As can be seen in
Table 36, past history and actions of individual fishers have a direct bearing on their ability to be
resilient, in the fact of fishery vulnerability. Interestingly, Skills/Education showed the highest
number of fishers’ comments. Those topics, however, seem to refer more to the ability of
individuals to obtain employment and income outside of fisheries, although that might also
indicate that they would be able to continue fishing part-time. In fact, social networks can be
seen as constituting the overall most important individual sources of resilience, given that
Fishing Experience and Sources of Income both refer to social networks which provide support
and allow individuals to continue their fishing enterprises. Ownership and Multiple Fisheries
both refer to socioeconomic settings in which individuals have additional security to enable them
75
to continue fishing as a livelihood. Entrepreneurship can be seen as related to Skills/Education,
reflecting personal qualities but also to the ability to see how individual action may contribute to
resilience and thus sustainability of the fisheries. Job Satisfaction provides position motivation
to continue fishing as a livelihood and thus constitutes resilience.
The two categories, FISHERIES and INDIVIDUALS, capture fisher knowledge and perspective
at very different scales. At the scale of the “fisheries within communities,” almost all of the
topics/codes referred to phenomena that represent vulnerability to the sustainability of the
fishery. In that sense, the fisheries in all of the communities are vulnerable, but not to the same
degree. Palacios shows the highest levels of vulnerability, followed by Seadrift, and then Port
O’Connor, Port Lavaca, and Galveston Bay.
Palacios, as noted, has been a major fishing community for several decades, especially in the
shrimp fishery. Like other shrimping communities, the fishers have faced increasingly difficult
circumstances during the past two decades, due to increased regulations, increased operating
costs, and the volume of imported shrimp and the consequent deflation of prices for wild-caught
shrimp in the U.S. The down-sizing of the fisheries in Palacios has been more obvious and the
impact on the local community more apparent than in any of the other communities, except
possibly for Seadrift (but which had smaller fisheries). Not unexpectedly, then, the fishers at
Palacios expressed the greatest amount of concern for the vulnerability of fisheries. Palacios had
the highest score of any community within the FISHERIES section on vulnerability (Table 33).
Palacios fishers were acutely aware of the issues within the fisheries that negatively affect
sustainability, and they talked about those issues more than in any other community. They were
less concerned, however, about topics within INDIVIDUALS that indicate resilience, especially
Skills/Education, in which they had the lowest total score. Given that few other options existed
for fishers and for the community, e.g., low tourism, gentrification, and industry, the decline in
the fisheries constituted critical vulnerability.
The once productive fishing community of Seadrift has seen its fisheries diminish substantially,
leaving the community in very difficult straits. The history, however, has been comparatively
recent, and the interviewees’ response can be interpreted in that light. Fishers remaining in
Seadrift were still concerned about conditions and events that could be seen as vulnerability,
again, scoring only behind Palacios. The historical point is that while Seadrift interviewees see
the community as very vulnerable, the reference actually may be to what little remains of once
vibrant blue crab and shrimp fisheries.
Based on the coding of the interviews, summary scores of the communities in terms of
vulnerability and resilience can be made. The ranking is presented in Table 37. The table is
constructed from the totals in Tables 33-36. Table 37 shows that all of the communities have
higher scores in FISHERIES for vulnerability than for resilience. The opposite is the case for
INDIVIDUALS.
76
Table 37: Summary scores of community vulnerability and resilience.
COMMUNITY
FISHERIES
INDIVIDUALS
VULNERABILITY
RESILIENCE
VULNERABILITY
RESILIENCE
372
603
792
1315
665
3747
749
17
0
10
23
20
70
14
78
30
70
29
40
247
49
296
499
468
370
255
1888
378
Galveston Bay
Port Lavaca
Seadrift
Palacios
Port O’Connor
TOTALS
AVERAGE
Even further, the FISHERIES and INDIVIDUALS scores can be combined to give a total
vulnerability and a total resilience score for each community. Those can also be ranked, as
shown in Table 38. As one would expect, Palacios shows the highest level of vulnerability;
Galveston Bay has the lowest. Port Lavaca has the highest level of resilience; Port O’Connor
has the lowest.
Table 38: Summary scores and community rank for vulnerability and resilience.
COMMUNITY
Galveston Bay
Port Lavaca
Seadrift
Palacios
Port O’Connor
AVERAGE
VULNERABILITY
450 (5)
633 (4)
862 (2)
1335 (1)
705 (3)
797
RESILIENCE
313 (4)
499 (1)
478 (2)
393 (3)
275 (5)
392
(7) Additional Information from the Interviews
The interviews and background historical notes compiled by the interns contained very useful
information that was often specific to the individual communities and not easily amenable to
coding across communities. The information and some remarks of its significance are presented
below.
The interviews were combed for information subsequent to the coding, and several items of
relevance emerged. One of those was levels of monthly income from fishing. Taken
collectively for all of the communities, the results indicate that in the majority of cases, 30 of the
48 instances in which information were available (62.5 percent), the earnings were $4,000 per
month or less. Only in six cases (12.5 percent) were the earnings in excess of $4,000 per month,
and in 12 instances (25 percent), they were $2,500 per month or less. While the majority of
incomes are above the poverty level, the incomes overall are not substantial, and they can be
taken to be an economic vulnerability. The degree of vulnerability increases when additional
77
interview information is taken into account, specifically the reports of 24 fishers, at least 33
percent, that operational costs exceeded their income levels.
Information of age-range was not available from all respondents in all communities, but
sufficient information was available to calculate average age across communities. In all of the
communities, more than 10 individuals reported an age of more than 55, giving a total of 49.
The total number of respondents was, again, 73, which means that even if all of those not
responding were below 55, which is unlikely, the percentage would be 49 or 73, or 67 percent.
The “graying” of commercial fishers is well known, and the individuals in the present study are
no exception, indicating community vulnerability.
Similar information to age-range can be seen in the number of years that fishers have been active
in commercial fishing. Again, a total of 49 individuals, 67 percent, indicated that they had more
than 20 years of fishing experience. Again, the percentage is likely higher. While the longevity
can be seen as an experiential indicator of resilience, the same number of fishers, 49, reported
that their participation in the industry has decreased, indicating lowered resilience.
Twenty-four of the respondents reported that they have children who have a tertiary (college or
university) level of education. Children with higher levels of education and thus opportunity and
income can count toward resilience, as more or less a social security safety net. In some cases,
fishers were insistent that they have not depended on their children and prefer never to be in that
position, even rejecting pleas from their children to accept help. It is doubtful, however, that the
availability of assistance is actually a form of resilience, since the assistance is typically not to
help with fishing but with life after fishing has ceased. In one moving instance, a fisher reported
that his adult sons, who had been helping him fish to the extent they could, confronted him and
told him that it was time to give it up, that he did not need to continue to work as hard as he did
and lose money in the process.
(B) Contextual Research
In many fishing communities throughout the United States, commercial fishing is negatively
impacted by the growth of tourism, recreation, recreational fishing, and gentrification, i.e., sociodemographic and socioeconomic changes in community make-up and characteristics. That is the
case in some of the communities in the research project. The degree and extent of the impacts
depends largely on how urbanized the communities are or what the potential for urbanization is.
To begin, the communities can be classified in terms of their economic dependence on fishing,
commercial and recreational. The focus here is specifically on the communities, not on the
fishing communities themselves. The assessment of community economic and social
dependence on commercial and recreational fishing is given in Table 39. The individual
communities of Galveston Bay, treated collectively in the interview analyses, are viewed here
separately.
78
Table 39: Community economic and social dependence on commercial and recreational fishing.
Community Rating and Comments
Bacliff
Commercial Dependence: Low. There are no commercial facilities in
Bacliff.
Recreational Dependence: Low. A few private docks are attached to houses
for recreational fishing or boating.
Social Dependence: Low. There are no significant fishing social activities.
Galveston
Commercial Dependence: Low. Galveston has a public dock for a
commercial fleet of finfish and shrimp boats. There are also wholesale buyers
and shippers. Commercial fishing is neither a major employer nor is it a major
source of revenue.
Recreational Dependence: Medium. Before Ike, there were numerous bait
camps, two long public fishing piers, beaches open to wade-fishing, charter
and head boats for hire. The city is dependent on nature based tourism, and
recreational fishing is part of that tourism but is not highly promoted.
Social Dependence: Low. Galveston tourism is based on two major strands:
its Victorian past, and cultivation of an ‘island feel’. The Victorian past model
promotes architectural reconstruction and festivals such as ‘Dickens on the
Strand’. The visual manifestation of ‘beach culture’ has been promoted,
however, by planting palm trees near the beach and by selling beach type
merchandise. There are numerous seafood restaurants that use a fishing motif,
but the city as a whole does not celebrate fishing culture in festivals or
visually.
San Leon
Commercial Dependence: High. Pre-Ike San Leon was home to two oyster
processing plants, two large oyster leaseholders, the Vietnamese crab fishery
and processing plant, and two Vietnamese shrimp docks with marinas.
Recreational Dependence: High. The community also has a large recreational
fishing center where both recreational and shrimp boats dock.
Social Dependence: High. The town’s slogan is “A small drinking
community with a large fishing problem.” A local landmark restaurant, written
up in out of state tourist articles, is a favorite gathering place for locals and
tourists.
Texas City
Commercial: Low. There is one wholesale shrimp and finfish dealer in Texas
City and an oyster processor. They existed there only because the city annexed
the unincorporated area in which they were located. Previously, this area was
part of San Leon and socially that is still the case. Several commercial
fishermen interviewed lived in Texas City but docked on Port Bolivar.
Recreational: Low. Pre-Ike, the Texas City Dike was home to five bait camps
and boat ramps. It also had some slips for shrimp boats. The dike was an
important recreational venue for family fishing. There was one abandoned bait
camp. There were also boat ramps on Dollar Bay. Despite Texas City’s being
a destination for recreational fishing, fishing did not appear to have a
significant economic impact on the community as a whole due to the
concentration of the petro-chemical industry.
Social Dependence: Low. Texas City is known as an industrial town, not a
fishing town. The Texas City Dike, however, was once a symbol of the town
79
Seadrift
Palacios
Port
O’Connor
and of family oriented recreational fishing. After the destruction of the bait
shops on the Dike after Ike, there were many reminiscences about days spent
fishing from the Dike on the local newspaper’s website.
Commercial Dependence: High. The Vietnamese crabbing community is
located in Seadrift as well as a shrimp fleet, but they have decreased
substantially in recent years. There is a commercial docking facility with two
buyers located.
Recreational Dependence: Low. Recreational fishing in Seadrift is relatively
low, due to essentially no infrastructure for tourism.
Social Dependence: Medium. At one time, social dependence was high.
Historically there was a large Vietnamese crabbing community, but this is now
in decline. The community also hosts a Shrimpfest, and houses are decorated
with fishing paraphernalia. Except for the remaining Vietnamese, the sense of
community appears to be in decline.
Commercial Dependence: High. Palacios is home to a large gulf shrimp
fleet, The Marine Education Center and TPWD Research Center. Commercial
fishing has been the major economic engine for the community.
Recreational Dependence: Low/Medium. The community hosts fishing
tournaments and fishing related festivals, and there is potential for growth of
recreational fishing. The relatively undeveloped infrastructure for tourism,
however, limits the importance of recreational fishing.
Social Dependence: High. Palacios hosts the “Shrimp o Ree,” Texas
Seafood, and Blessing of the Fleet festivals. The shrimp fleet is networked
through a few major families who own fleets and provide loans and other
support to local fishers. The Vietnamese community is networked through a
local church.
Commercial Dependence: Low. Port O’Connor had a bay and bait fishery,
but they are in rapid decline.
Recreational Dependence: High. The community is highly dependent on the
sport fishery. It hosts several large fishing tournaments. One of the founders of
CCA has his summer home there. There are a number of bait camps, boat
sheds and ramps. New resort style residential compounds are being built that
include boat ramps and slips.
Social Dependence: High. Recreational fishing is promoted in various ways
throughout the community, specifically fishing tournaments and merchandise
in shops (for sale even in the ice cream shop). There is also a Fisherman’s
Chapel. The major hotel/restaurant complex is now being remodeled to attract
tourists for recreational fishing.
The information and assessments can be summarized in tabular form, as shown in the following
Table 40.
80
Table 40: Summary of commercial, recreational, and social dependence by community.
COMMUNITY
Bacliff
Kemah/Seabrook
Galveston
San Leon
Texas City
Port Lavaca
Seadrift
Palacios
Port O’Connor
COMMERCIAL
Low
Medium
Low
High
Low
Low
High
High
Low
RECREATIONAL
Low
Low/Medium
Medium
High
Low
Medium
Low
Low/Medium
High
SOCIAL
Low
Low
Low
High
Low
Low/Medium
Medium
High
High
As the table shows, communities with high dependence on commercial or recreational fishing
will have high social dependence. Communities in the most diversified economies or in
industrial towns will have low or low/medium social dependence.
Community socioeconomic vulnerability can also be assessed using historical and contextual
information, again focused on the communities and not solely on fishing. The results are given
in Table 41.
Table 41: Community Socioeconomic Vulnerability.
Bacliff
Vulnerability: Moderate. Bacliff is primarily composed of older homes
and is a residential community with no large industry. It is near to the
petrochemical facilities in Texas City, thus potentially exposed to a
technological disaster. Given that Bacliff is not very dependent on
fisheries, new or expanded regulations would not have a significant impact
on the well-being of the community as a whole.
Galveston
Vulnerability: High. Galveston has a high proportion of people living in
poverty. Its housing stock is old, and as a barrier island it is prone to
natural disasters. Galveston also depends on a few key resources
economically, and it is very vulnerable to economic perturbations.
Galveston is not very dependent on fisheries, and thus regulations would
not have a big impact on the well-being of the community. However, due
to Galveston’s other vulnerabilities, displaced fishermen would have a
difficult time.
Kemah/Seabrook Vulnerability: Low. Kemah and Seabrook are primarily residential areas
with the usual complement of shops, small businesses, superstores, and
hotels that serve residents and visitors. There is no major industry in either
city. The towns are primarily bedroom communities for commuters who
work in nearby industry, the university, NASA or myriad businesses in
downtown Houston. These communities are not very dependent on
fisheries, and regulatory changes would have only a moderate impact on
the communities as a whole. A diversified economy surrounds these
communities. There are several industrial parks, a major port, and
numerous government contractors within a 30 minute drive.
81
San Leon
Texas City
Port Lavaca
Seadrift
Palacios
Port O’Connor
Vulnerability: High. San Leon is unincorporated. It has a large
Vietnamese community, many of which are not fluent in English. The
housing stock is mixed. There are large substantial houses along the bay,
trailers and older recreational houses that are now permanent homes. The
fishing industry is an important part of the economy to San Leon, and the
decline of shrimp and crab has impacted the town. A majority of its
housing stock was severely damaged during Ike and the town is still
struggling.
Vulnerability: Medium. Texas City has a seawall that mitigates the
impact of storm surges from hurricanes. It has a mixed stock of houses
ranging from the early 1900s to very modern, and it is a diverse community
with a range of ethnic and income groups. It is home to one of the oldest
petrochemical complexes on the coast and was the site of a major disaster
in the 1940s, and there have been several explosions since. Its proximity to
these plants make it vulnerable to technological disaster.
Vulnerability: Low. Port Lavaca has a diversified economy that is a mix
of mercantile, heavy industry, tourism, shipping and agriculture. Although
its infrastructure is vulnerable to storms, as a community it has a
diversified economy and population and could most likely rebound from a
hurricane. It is not dependent on the fishery for a significant portion of its
revenue and hence regulation would not severely harm the community as a
whole. There are several chemical facilities close to Port Lavaca and thus
the potential for exposure to technological disaster exists.
Vulnerability: High. Seadrift is unincorporated. It has a large Vietnamese
community, many of which are not fluent in English. Seadrift has a legacy
of conflict between Vietnamese and Anglo fishermen. The commercial
fishery is in economic trouble, and a recreational fishery is not well
established. There is one industrial facility across the bay from Seadrift
and several more within an hour’s drive.
Vulnerability: High. Several families own most of the commercial fleet
and employ both local and migrant labor. Dense social ties bind the
members of the fishing community together. There is a large Vietnamese
community in Palacios that is not fluent in English and unable to find
employment outside of the fishery. There are some industrial opportunities
within driving distance, but outside of the town. Many fishermen
interviewed stated a reluctance to work in these plants, however, for fear of
exposure to pollutants. Because Palacios is still dependent on fishing, it is
vulnerable to regulatory changes and storms.
Vulnerability: High. Port O’Connor has few remaining commercial
fishers, due to the combination of imports and gentrification. It has always
been primarily a sport-fishery oriented community, but with gentrification,
even the bait shrimpers are finding it difficult to find dock space or sell
their catch. The recreational fishery is the economic driver of the
community; hence regulations aimed at the recreational sector would harm
the community.
82
Community socioeconomic vulnerability depends essentially on economic diversification. The
communities with the greatest amount of diversification have the lowest vulnerability, and the
communities dependent primarily on commercial fishing, recreational fishing, or heavy industry
show high vulnerability.
Gentrification typically is involved in the transformation of communities as they undergo
economic growth and diversification. Commercial fishing is often negatively impacted
economically, due to increases in coastal land value and use and accompanying marginalization
of local residents. Information on gentrification was collected for the nine communities, and the
rate and amount of gentrification can be qualitatively assessed. That information is presented in
Table 42. Note should be made that the relationship between gentrification and vulnerability is,
however, not entirely straightforward. Virtually no gentrification means that a community will
not have much economic diversity and will thus suffer vulnerability. Similarly, very high levels
of gentrification may mean that commercial fishing is heavily marginalized. More moderate
levels more likely represent instances of economic diversification, which indicates resilience.
The relationship is likely u-shaped.
Table 42: Levels of gentrification in the nine coastal communities.
Bacliff
Low. Expensive houses are beginning to dot the shoreline, but most homes
and restaurants are still modest. Bacliff is unincorporated, which may have
stemmed gentrification due to lack of control as to what can be built
adjacent to an exclusive subdivision. Urban sprawl has already occurred
around Bacliff, but the in-migration of amenity seekers is not yet
substantial. Bacliff, however, has considerable open space which could be
filled with houses if there were substantial in-migration.
Galveston
Medium. Gentrification has occurred near the medical center, in the
historic district, and in the historical downtown area. Overall, however, the
housing stock in Galveston is old, and there has not been new building
behind the sea wall. Many people who work in Galveston live on the
mainland because of this. City Hall is concerned that Galveston is losing
its middle class and becoming a divided city with the wealthy second home
owners on the west of the island, a few well-to-do in the historic district,
and the poor on the east end and middle portions of the island. Few people
are moving to the island. Amenities draw second-home owners along the
west end of the island, and this is the only place undergoing sprawl.
Kemah/Seabrook High. Commercial docking spaces have largely been replaced by
recreational ones. Clear Lake, on which both Kemah and Seabrook are
located, has one of the highest concentrations of recreational slips in the
nation. This area became suburbanized years before land prices began to
rise substantially. Now, prices for houses on the bay have risen, and both
towns are bedroom communities for other areas. A few blocks from the
bay, however, prices are normal for the region. If Seabrook follows its
master plan, its waterfront will gentrify but will retain a working
waterfront.
83
San Leon
Texas City
Port Lavaca
Seadrift
Palacios
Port O’Connor
Low. There are a few gated communities close to the shore but for the
most part, the town has not changed substantially. Like Bacliff, San Leon
is unincorporated and has no zoning. Also like Bacliff, San Leon has
considerable open space and if there were substantial in-migration, new
houses could be readily built.
Medium. Texas City is building several new subdivisions but these are
not amenity based. The previous mayor had a beautification program, and
several initiatives were launched at that time. As League City fills in open
space, it is possible that Texas City will expand its subdivisions.
Medium. The downtown area is undergoing revitalization, with old
buildings being re-fitted for new uses, such as shops that cater to tourists
and local businesses that cater to locals. The commercial docks are no
longer fully functional, and two of them are now full of recreational boats.
Only two docks remain where commercial boats can land. There are gated
communities to serve the managerial class at the local plants. Beach-front
land values are rising.
Low/Medium. Like Port O’Connor, Seadrift is on the cusp of
gentrification. There are two new gated developments, and land prices are
increasing. As Port O’Connor land prices go up, developers are looking to
expand near Seadrift. The community currently, however, has less
infrastructure and will likely develop more slowly.
Low. The port is being re-vitalized, but it is unclear how this will impact
the commercial boats now using it. There are plans for a recreational boat
building facility at the port and some concern that commercial boats will be
pushed out. Palacios has only limited tourism.
Medium. At the time of the field research, only two docks existed where
commercial boats could sell their catch, but one of those has now converted
to a recreational dock. Older ranches and farms are being sold for large
gated communities. Land prices have recently begun to skyrocket, and
taxes on some properties have increased. Although the real estate market
for land is growing rapidly, new house construction has only barely begun.
The potential for gentrification is high but is not yet manifested at that
level.
(C) Summary: Ground-Truthing Results
An interview protocol was developed and administered to nine coastal communities on the Texas
coast, five communities in the Galveston Bay Complex, three communities on San Antonio Bay,
and one community on Matagorda Bay. A version of the protocol, modified for gentrification,
was administered to bankers, developers and elected officials in the city of Galveston. A total of
106 interviews were conducted in the nine communities. A sub-set of the interviews were
analyzed to obtain keywords, essentially names of topics on which the interviewees responded.
A keyword coding sheet was developed, containing approximately 70 topics. In addition, the
topics were separated into two major sections, FISHERIES and INDIVIDUALS, according to
84
whether the focus of the topic was on the larger aggregate of fishery or more specifically on
individual fishers.
The responses in the 106 interviews were entered into the coding sheets, providing raw scores of
the numbers of individuals by community who commented on or talked about the topic (Table
29). For purposes of more accurate comparison of comments across communities, the total
number of comments by community was divided by the total number of interviews, giving a rate
measure (Table 30). Comparisons were of each of the 70 keyword/codes and for each of the
higher-order categories to which sub-sets of the codes were assigned. Totals for each of the two
sections, FISHERIES and INDIVIDUALS, were compiled, and comparisons were made by
community. The community that scored the highest on FISHERIES was Palacios, and the lowest
score was for Galveston Bay. On the section INDIVIDUALS, the community with the highest
score was Seadrift, and the lowest score was for Port O’Connor.
The conclusions to this point were that the two communities where commercial fishing is the
most focused within the community, Palacios and the Galveston Bay, showed the most concern
for topics relating to FISHERIES and to INDIVIDUALS. The least concern was shown in the
two communities, Port Lavaca and Port O’Connor, in which commercial fishing was the least
focused and least historically salient.
A more fine-grained analysis of the interview results was carried out in relation to vulnerability
and resilience. Again, following the division into the sections FISHERIES and INDIVIDUALS,
qualitative decisions were made as to which of the topics reflected fishery vulnerability and
which ones reflected resilience. The decisions were not always completely clear-cut, but most of
them were straightforward. Once the decisions were made, the total number of respondent
comments per topic were calculated and summed into the higher order categories and by
communities. The totals for vulnerability far exceeded those for resilience, by several
magnitudes (Tables 33 and 34). The topics receiving the most comments and thus concern were
Infrastructure, Land/Water Use Changes, and Catch Levels, in that order.
For INDIVIDUALS, the number of comments was more evenly divided between vulnerability
and resilience, although the totals are comparatively greater for the latter (Tables 35 and 36).
The topics indicating vulnerability were Sources of Income (Fishing only, No social security),
Changes in Liquidity (All assets in fishing), and Fishing Experience (First generation). The
community with the highest total score was Galveston Bay. For resilience, the topics were
Skills/Education, Fishing Experience (more than first generation), Sources of Income (more than
fishing only), and Multiple Fisheries. The community with the highest total score was Port
Lavaca.
In order to provide a more comprehensive picture, the results for vulnerability in FISHERIES
and INDIVIDUALS were combined, as were those for resilience, and the communities were
ranked in terms of the totals (Tables 37 and 38). Again, the results for vulnerability were
85
Palacios with the highest score and Galveston Bay with the lowest score. The results for
resilience were Port Lavaca first, followed closely by Seadrift, and with Port O’Connor last.
Additional information that indicated vulnerability or resilience was gleaned from the interviews.
On the vulnerability side were low monthly incomes, some below the poverty level, and the age
of the fishers (over 55). On the resilience side was years of experience in fisheries and number
of children with college or university levels of education.
Also, each of the nine communities was described qualitatively in terms of the vulnerability due
to economic dependence on commercial fishing, recreational fishing, and social dependence on
the fisheries (Table 39). The results are summarized in Table 40, indicating (1) that communities
with high dependence on commercial or recreational fishing will have high social dependence on
fisheries, and (2) that communities with the most diversified economies or within industrial
towns will have low or low/medium social dependence.
Lastly, qualitative accounts were provided of the vulnerability of the nine communities due to
economic dependence (Table 41) and gentrification (Table 42). Communities within the least
diversified economies scored the highest in vulnerability, and vice-versa. As noted above, very
low and very high levels of gentrification likely indicate vulnerability for communities, whereas
moderate levels may indicate economic diversification and thus resilience. The nature of the
relationship may ultimately depend on how gentrification and fisheries actually interact.
B. Significant Problems
There were no significant problems. Collection of ethnographic (ground-truthing) data proved
to be more difficult than anticipated in the Galveston Bay communities, due to the distribution of
the fishers throughout the communities. Interviewing was necessarily by dock intercept. The
major consequence of the problem was the collation of the data from the five communities in one
Galveston Bay “community” for purposes, thereby obscuring differences among those
communities.
C. Need for Additional Work
Despite the difficulties with obtaining expected data sets in the Galveston Bay communities, the
ground-truthing was effective, confirming the validity of the social indicators proposed. No
additional work is needed.
86
VII. EVALUATION
A. Extent to Which Project Goals were Met
1. Evaluation Strategy
The purpose of this project was to develop and evaluate social indicators based on secondary
data to measure the concepts of dependence, gentrification, vulnerability and resiliency. The
evaluation strategy relied upon two simultaneous and independent processes. The social
indicators were developed concurrent with field work. Though the field work was described as
“ground-truthing,” in fact it was a grounded emergent process of discovery of the concepts of
dependence, gentrification, vulnerability and resiliency as they relate to each study site. Groundtruthing would be the opposite, taking concepts as defined and verifying them in the study sites.
In terms of external validity and triangulating the concepts the process we used was much more
rigorous but posed many difficult issues of assessment which we will detail later. We will begin
with a description of the differences in the processes of the social indicators approach and the
ethnographic approach.
(A) Description of the Social Indicators Process
The development of the social indicators was deductive in nature. We began with a thorough
literature review of the theoretical constructs and concepts of dependence, gentrification,
vulnerability and resiliency. Next we developed hypotheses about the nature of the relationships
of the indicators to the concepts. We created the measures and made observations reported in the
rankings tables. Last we made generalizations about the place in the form of assessments on
each concept that will appear later in this chapter.
The social indicator process was quantitative; relying solely on secondary numerical data. There
were no field visits and no interviews with local residents in this process. The information used
consisted entirely of empirical numerical observations. As such, the indicators lost some of the
context of place and the complexity that exists in the local milieu. However, it was possible to
develop indicators for over 100 places in a relatively short time period. In addition, it required
only the work of one researcher. Quantitative measures usually have a high degree of reliability.
In this case the constructs were measured identically in all communities. However, external
validity can suffer in these circumstances, in as much as the constructs or concepts may not
relate to actual community conditions.
The development process for the social indicators was also nomothetic in nature. It sought to
develop indicators that were generalizable to all coastal or coastal adjacent communities in
regard only to the specific concepts of interest. It sought to combine information on areas of
agreement across all communities to develop commonality in the concepts. As such, it
downplays the differences, often treating them as errors or outliers, a sacrifice made so that
87
generalizability can be achieved. Another drawback is that reducing community processes to a
few key measures can obscure reality in a reductionist simplification. All of the above criticisms
are not particular to this research but part of the general critique of normative science that is
typically deductive, quantitative, and nomothetic.
(B) Description of the Ethnographic Process
The ethnographic process by definition was inductive. It began with observations based upon
open-ended interviews with key informants. In this sense the ethnographic process fleshed out
the concepts of dependence, gentrification, vulnerability and resiliency in a grounded emergent
process based on the reality of the interviewees. These observations were turned into summaries
and then generalizations that appeared in the Summary of Ground-Truthing Results (Section VI,
Section 2C).
The ethnographic process is also qualitative. The focus of this research is descriptive of each
community site. The data in this process consisted entirely of words based on interviews and
personal observations. A major drawback to this process is that it is time consuming, expensive,
and offers difficulty in both analysis and interpretation. Three full-time field workers collected
data in nine community case study sites for three months. These field workers were overseen by
three Ph.D. level part-time investigators. In addition many weeks were spent by the investigators
coding and analyzing the data. The result of this effort was a very detailed description of nine
communities. A major advantage of a qualitative case study approach is that it has a high degree
of external validity. That is to say the results reflect what is actually happening in the
community. However, there may be issues with internal validity in as much as the link to the
construct or concept of interest may not have been measured in a repeatable, reliable fashion.
The ethnographic process certainly represents idiographic research. The purpose of idiographic
research is to completely understand concepts and mitigating factors in context. The emphasis
on complete understanding implies intensive observation and data collection. As such relatively
few places can be studied and therefore generalizability to broader populations is sacrificed.
However, idiographic case studies excel at providing converging emergent evidence of
constructs.
(C) Differing Processes with a Converging Reality
There is a tension in research between validity and reliability. Quantitative research tends to be
more reliable in that it uses repeated measures in a consistent way. This is less true of qualitative
research which tends to build constructs from emergent findings. Since many places differ, the
processes often vary greatly. Qualitative research on the other hand tends to have greater
external validity. That is to say the construct is grounded firmly in the real world. Quantitative
research tends to simplify or reduce constructs into easily measured pieces that may not
accurately reflect real world conditions. In our project we decided to do both to see how the
quantitative and qualitative results vary.
88
A key concept in the measurement of constructs is the interchangeability of indicators. Any
reasonable indicator of a construct should be correlated to other indicators of that same construct.
In other words if they are both measuring the same thing they should also strongly relate to one
another in the same directional pattern. The degree that two differing indicators of the same
construct relate to reality is usually referred to as construct validity. To establish construct
validity, ideally the quantitative data in our project should be highly correlated with the
qualitative data, which best reflects the objective conditions in the community.
(D) Interrater Reliability
Interrater reliability is the degree to which independent observers evaluate the characteristics of a
subject and reach the same conclusion (Lombard et al., 2002). High level of agreement in
ratings generally reflects the reliability of the standards and process. This is true especially if
two different raters are applying the same criteria and reaching the same results. However in this
case there are two completely differing sets of criteria and processes. Here a high level of
agreement reflects convergence of a construct with reality. In other words rather than being a
reflection reliability (receiving the same results from repeated measures using the same criteria)
it is a reflection of both construct and external validity (the link between a construct and
observed reality). For this use we agree with Lombard et al. (2002) who have argued
convincingly that a more accurate term would be interrater agreement.
There are some widely reported measures of agreement used to assess interrater reliability or
agreement. They are percentage agreement, correlations based indicators such as Pearson’s r or
Spearman’s rho, and Cohen’ kappa. Each of these measures has some significant advantages and
drawbacks but taken together they allow for a more complete assessment of interrater agreement.
Percentage agreement is easily understood and has a straight forward interpretation but can be
misleading. The percentage agreement is often inflated because a portion of agreement could be
directly due to random matching. This is especially true with constructs with relatively few
categories. For example, with three categories up to 11.1% (.333*.333 or 1 in 3 * 1 in 3) of
agreement could be due to random matching. Correlational techniques measure covariation but
not the extent in which there is identical agreement in the categories. Bivariate correlations are
generally interpreted in analysis to be substantial above 0.6. The last statistic used is Cohen’s
kappa that measures interrater agreement and ranges from -1.0 to 1.0. The closer the number is
to 1.0 the greater agreement is above and beyond random matching. If the number is
approaching zero then the level of agreement is close to what would be expected by chance. If
the number is below zero and approaching -1 then the agreement is less than what would be
expected by just random matching.
Cohen’s kappa is calculated by taking the percentage of agreement [Pr(a)] and subtracting the
probability of random agreement [Pr(b)], divided by one minus the probability of random
agreement [Pr(a)-Pr(e) / 1-Pr(e)]. The probability of random agreement is calculated by dividing
1 by the number of categories for rater one and multiplying it by 1 divided by the number of
89
categories [Pr(e) = 1/k * 1/k where k = the number of categories for the rater]. Cohen’s Kappa is
generally interpreted with the following framework from Landis and Koch (1977): less than zero
= no agreement; 0 to 0.20 = slight agreement; 0.21 to 0.40 = fair agreement; 0.41 to 0.60 =
moderate agreement; 0.61 to 0.80 = substantial agreement; and 0.81 to 1.0 = almost perfect
agreement. A t value can be calculated for kappa by dividing the kappa value by the asymptotic
standard error when the null hypothesis is true (the true value is 0). This t value has an
associated statistical probability that is often reported.
While both percentages and correlation techniques tend to be liberal and over-assess levels of
agreement, Cohen’s kappa is considered a very conservative measure and underestimates the
strength of agreement. This is because it only includes exact matches as agreement, when often
misses are only a category off and the raters are actually in relative agreement. It is possible to
use a weighted kappa statistic to account for close misses but it is not commonly done and the
statistic is not included in any major software packages.
(E) Coding Issues for the Secondary Data Indicators
To ensure content validity with the constructs of dependence, gentrification, and
vulnerability/resiliency multiple indicators of each were developed. Specifically for fishing
dependence there were three indices: 1) commercial, 2) recreational, and 3) social. For
gentrification there were three indices: 1) urban sprawl, 2) natural resources migration, and 3)
retirement migration. Last for vulnerability/resiliency there were seven indices: 1) population
composition, 2) poverty, 3) housing characteristics, 4) labor force, 5) natural and technological
disasters, 6) housing disruptions, and 7) personal disruptions.
To evaluate the agreement of the social indicators with the ethnographic research it was
necessary to code the indices into the same categories employed in the qualitative analysis.
These categories were: 1) low, 2) medium, and 3) high. Each separate community (N=125) was
coded into one of the thirds (low, medium, or high) based on the index factor score, so the
response categories within the nine communities are not evenly distributed (e.g. 3 lows, 3
mediums, and 3 highs). For the dependence indices of commercial, recreational, and social,
direct comparisons for the ratings for the social indicators and the ethnographic data could be
made. This is because in the ethnographic coding process the domains of commercial,
recreational, and social dependence emerged from the content analysis. However for the
gentrification and vulnerability/resiliency ratings the ethnographic analysis only examined the
central constructs of gentrification and vulnerability/resiliency. This is not to say that the
ethnographic research failed to achieve content validity. The content analysis revealed the full
dimensions of gentrification and vulnerability/resiliency. However, the dimensions were most
important in particular communities and sometimes did not emerge at all in other communities.
As such the analysis produced just a single rating for each for gentrification and
vulnerability/resiliency. This posed a particular challenge for the social indicators as there were
multiple indicators for each construct. To evaluate rater agreement it was necessary to condense
90
the multiple indices for gentrification (three indicators) and vulnerability/resiliency (seven
indicators) into single assessments.
Ideally two single indices measuring gentrification and vulnerability/resiliency could be
constructed to cover all the dimensions of the concept and still remain unidimensional.
However, this could not be achieved with a satisfactory level of reliability and unidimensionality
within the principal components analysis. The next effort involved summing the factor scores
into a single score and then placing the scores into thirds to match the ethnographic categories.
This was done and produced highly consistent results with the ethnographic data. However, such
a summing of factor scores is seriously inadvisable since each index is assumed to be of equal
importance in the analysis, which is not likely to be the case. Several weighting schemes were
also attempted via factor and canonical correlation analysis. All produced highly similar results,
varying very little from the simple summation of the factor scores but adding a layer of
complexity that was hard to analyze and discuss. Ultimately all efforts at summing the factor
scores were abandoned due to methodological problems.
To produce a single score, a simple modal response coding scheme was employed. In this case
we simply added up the number of high, medium, or low categories and the category that
occurred most frequently within a community was assigned to that community. In several cases
there was a tie between the low and high categories, when that occurred the medium response
was assigned. This coding scheme produced results highly correlated with the summing schemes
described above however it was not subject to the same methodological shortcomings. Since the
categories were treated at the nominal level there were no issues with the weighting of indices.
Additionally the interpretation of the results is simple and straightforward. Last, such an
approach has obvious face validity and produces results consistent with much more complicated
procedures but with methodological flaws.
(F) Interrater Agreement Results
Table 43 presents the results of the interrater agreement analysis for commercial, recreational,
and social fishing dependence. For the commercial fishing dependence measure the quantitative
and ethnographic assessments matched in 66.67% of the communities. This produced a
Spearman’s rho of 0.589 however it was not statistically significant due to the small sample size
(n = 9 cases). Cohen’s kappa was 0.500 and was statistically significant, reflecting a moderate
level of agreement between the two techniques. In two of the three mismatches the categories
were only off by one category. However, in Port O’Connor the mismatch was off by two full
categories. The quantitative results indicated high commercial dependence based on the 2007
value and pounds of landings and numbers of commercial licenses and dealers.
The
ethnographic research was conducted in 2009. It is possible that the conditions of the
commercial fishing industry in Port O’Connor changed greatly in the two year delay in data
reporting. It is also possible that the ethnographic results were underestimated due to selection
bias in key informants. In either case, in this community there was a large difference in ratings.
91
Table 43: Commercial, Recreational, and Social Dependence.
Commercial Dependence
Community
Port Lavaca
Recreational Dependence
Quantitative
Ethnographic
Differing
Quantitative
Ethnographic
Differing
Index
Assessment
Classification
Index
Assessment
Classification
*
Medium
Medium
Medium
Low
Seadrift
High
High
Port O'Connor
High
Low
Palacios
High
High
Seabrook
Medium
Medium
San Leon
Medium
High
Low
*
*
Low
High
High
Medium
Medium
Medium
Medium
High
High
Galveston
Low
Low
Medium
Medium
Texas City
Low
Low
Low
Low
Bacliff
Low
Low
Low
Low
Matched on 6 of 9 Communities
Matched on 9 of 9 Communities
66.67% Matched
100% Matched
Spearman's rho .589, P. .095
Spearman's rho 1.0, P .000
Kappa .500, P .022
Kappa 1.00, P .000
Social
Dependence
Community
Quantitative Index
Port Lavaca
Ethnographic
Assessment
Medium
Medium
Seadrift
Port
O'Connor
High
Medium
High
High
Palacios
High
High
Seabrook
Medium
Low
San Leon
High
High
Galveston
Low
Low
Texas City
Low
Low
Bacliff
Low
Low
Differing
Classification
*
*
Matched on 7 of 9
Communities
77.78%
Matched Spearman’s rho
.900, P.000
Kappa .660,
P
.005
Continuing with Table 43 but examining the results for recreational dependence, we found
complete agreement between the quantitative and qualitative results. They matched on all nine
communities producing 100% agreement, a Spearman’s rho of 1.0, and a Cohen’s kappa of 1.0.
In the case of recreational dependence the techniques were in perfect agreement.
Also on Table 43 the interrater agreement results for social dependence matched on seven of
nine communities (77.78%). Spearman’s rho for the two techniques was 0.900 and this was
statistically significant. Cohen’s kappa was 0.660 and was statistically significant and reflects
substantial agreement between the raters. In the two mismatched communities the categories
were off by only one category, with the quantitative ratings higher than qualitative.
92
Table 44 presents the interrater agreement for the gentrification construct. For gentrification six
of nine communities matched (66.67%). Spearman’s rho was 0.837 and was statistically
significant. Cohen’s kappa for gentrification was 0.500 and was statistically significant,
reflecting moderate levels of agreement. It should be pointed out that the correlation for
gentrification was substantially higher than for commercial fishing dependence (0.837 compared
to 0.589) even though both had exact same percentages of agreement (66.67%) and kappa was
0.500 for both analyses. This illustrates the importance of using multiple statistics to assess
agreement. It shows that in this case gentrification actually is a closer match in agreement as
seen in the Spearman’s rho. For gentrification, the three mismatch categories were only off by
one category and in every case the quantitative results rated higher and the qualitative results
were lower. This produced high levels of covariability and thus a higher correlation. For
commercial fishing dependence there was one mismatch off by two categories, and for the other
two mismatches, in one case the quantitative rating was high and the other low. This lack of
patterning, seemingly random, reduced covariability and the overall correlation and reflects
issues of reduced reliability between the two processes. In summation, though the levels of
agreement as seen in the percentages and kappa are identical, in fact the level of agreement in
gentrification is higher than it may first seem.
Table 44: Quantitative Gentrification Indices.
Quantitative Gentrification Indices
Community
Port Lavaca
Seadrift
Port
O'Connor
Palacios
Seabrook
San Leon
Galveston
Texas City
Bacliff
Urban
Sprawl
Low
Low
Natural Resources
Migration
High
Low
Retirement
Migration
Medium
High
Modal
Response
Medium
Low
Ethnographic
Assessment
Medium
Low
Differing
Classification
High
Low
High
Medium
Medium
Low
Medium
Medium
Low
High
Medium
High
High
Low
High
High
Low
Medium
Medium
High
Low
High
Low
High
Medium
Medium
High
Low
Medium
Low
High
Low
Medium
Medium
Low
*
*
*
Matched on 6 of 9
Communities
66.67% Matched
Spearman's rho
.837, .005
Kappa
.500,
p.024
Table 45 presents the interrater agreement analysis for vulnerability/resiliency. In this case the
two techniques matched on seven of nine communities (77.78%). Spearman’s rho for
vulnerability/resilience was 0.608 but was not statistically significant due to the small sample
size (n = 9). Cohen’s kappa was 0.625 and was statistically significant and reflected substantial
agreement between the techniques.
93
Table 45: Vulnerability/Resilience Indices.
Quantitative Vulnerability/Resiliency Indices
Population
Housing
Community
Composition
Poverty Characteristics
Port Lavaca
High
High
Seadrift
High
Port O'Connor
Labor Natura/Techno Housing
Personal
Qualitative
Modal
Ethnographic
Force
Disasters
Low
High
Low
Low
Low
Low
Low
High
Low
High
Low
High
Low
High
High
High
Medium
Low
High
Low
Low
Medium
Low
High
Palacios
High
High
Low
High
Low
Low
High
High
High
Seabrook
Low
Low
Medium
Low
High
High
Low
Low
Low
San Leon
Medium
High
Low
High
High
Low
High
High
High
Galveston
High
High
Low
High
High
Medium
High
High
High
Texas City
High
Medium
Medium
High
High
Low
Medium
Medium
Medium
Medium
High
Low
High
High
Medium
High
High
Medium
Bacliff
Matched on 7 of 9 Communities
Disruptions Disruptions
Quantitative
Response
Differing
Assessment Classification
*
*
77.78% Matched
Spearman's rho .608, P .082
Kappa .625, P .005
2. Overall Evaluation of Agreement
Each of the goals in the four objectives were attained. In general, the agreement between the
social indicators and the ethnographic techniques was very strong. The weakest level of
agreement was seen in commercial fishing dependence where there was a two category
mismatch and the other two misses were not in any pattern. Still the level of agreement by any
objective standard for that construct would still be considered moderate, with six of nine
communities in agreement and two of three mismatches off by only one category. The second
weakest level of agreement was seen in the gentrification construct but the three mismatches
were all by one category and were all in the same pattern of the quantitative results rating high
and the qualitative results rating low. The remaining constructs of recreational and social fishing
dependence and vulnerability/resiliency all exhibited substantial to perfect levels of agreement.
Although the differing techniques produced some mismatched outcomes, in general the
differences seem to be small. Both techniques show high levels in agreement in classifying
communities along these constructs.
3. Dissemination of Results
Copies of this project’s Final Report will be published and distributed to various federal and state
fishery agencies, university extension/Sea Grant offices, and industry associations. In addition,
PDF copies of the Final Report will be made available for download from the Foundation’s
website under Foundation Research (www.gulfsouthfoundation.org).
94
Summary reports of the project’s findings will be published as part of the “Foundation Project
Update” section of the “Gulf and South Atlantic News,” a publication of the Gulf & South
Atlantic Fisheries Foundation, Inc. This newsletter is distributed to over 300 organizations and
individuals throughout the region. An electronic version of this newsletter (PDF) is also
included in the regular updates to the Foundation’s website.
Copies of the Final Report will be made available for download from the HARC website
(http://www.harc.edu/). Additionally, HARC creates lay language websites for each of its
projects. It has a large network of natural resource partners, and notifies them via e-mail,
advising them the report will be sent to them, with a link to it. HARC will also advertise the
report through its newsletter. HARC participates in the “State of the Bay Symposium” for
Galveston Bay and will present results in that venue.
VIII. RECOMMENDATIONS
The social indicators identified in this project can be applied to coastal communities as a method
to assess community vulnerability, resilience, sustainability, and well-being. They include: (1)
social, economic, ecosystem and natural systems, social disruption, and gentrification; and (2)
natural resource dependence and well-being factors. The latter set of factors included fishing
dependence factors related to economics (commercial fishing, recreational fishing, and nonconsumption values) and society (institutions, cultural heritage, and landscape and built
environment).
95
IX. REFERENCES CITED
AngelouEconomics. 2008. Galveston Island Tourism and Economic Analysis. Final Report.
Abel, Nick, David H. M. Cumming, and John M. Anderies. 2007. Collapse and Reorganization
in Social Ecological Systems: Questions, Some Ideas, and Policy Implications. Ecology
and Society. 11(1): 17 (online).
Acheson, James M. 1981. Anthropology of Fishing. Annual Review of Anthropology. 10: 275316.
Adger, Neil. 2000. Social and Ecological Resilience: Are They related? Progress in Human
Geography. 24(3): 347-364.
Adger, W. Neil, Terry P. Hughes, Carl Folke, Stephen R. Carpenter, and Johan Rockstrom. 2005.
Social-Ecological Resilience to Coastal Disasters. Science. 309: 1036-1039.
Ahmed, Atiq Kainan, 2006. Concepts and Practices of "Resilience": A Compilation from
Various Secondary Sources. A Working Paper Prepared for Coastal Community
Resilience (CCR) Program. U.S. Agency for International Development. US IOTWS
Program Document No. 05-IOTWS-06. Bangkok, Thailand: U.S. Indian Ocean Tsunami
Warning System (IOTWS) Program.
Allen, John C. 2007. Morphing Rural Community Development Models: The Nexus between the
Past and Future. Community Investments. 19(1): 16-30.
Armor, David J. 1974. "Theta Reliability and Factor Scaling." In Herbert L. Costner. (ed.).
Sociological Methodology. San Francisco: Josey-Bass.
Armstrong, Paula S., and Michael D. Schulman. 1990. Financial Strain and Depression among
Farm Operators: The Role of Perceived Economic Hardship and Personal Control. Rural
Sociology. 55(4): 475-93.
Atkinson, Rowland. 2000. Measuring Gentrification and Displacement in Greater London.
Urban Studies. 37(1): 149-165.
Aulds, TJ. 2009. Dike’s return hindered by cost, amount of damage. Galveston Daily News.
Jan. 6. Web access. http://galvestondailynews.com/story.lasso?wcd=132155. accessed
March 20, 2009.
Bell, Michael M. 1992. The Fruit of Difference: The Rural-Urban Continuum as a System of
Spatial Identity. Rural Sociology. 57(1): 65-82.
Belyea, Michael J., and Linda M. Lobao. 1990. Psychosocial Consequences of Agricultural
Transformation: The Farm Crisis and Depression. Rural Sociology. 55(1): 58-75.
96
Blount, Ben G. 2002. Keywords, Cultural Models, and Representation of Knowledge: A Case
Study from the Georgia Coast (USA). Occasional Publication, Number 3. Athens, GA:
Coastal Anthropology Resources Laboratory, Department of Anthropology, University
of Georgia.
Blount, Ben G. 2006. Factors Affecting Participation in Marine Fisheries: Case Studies in
Georgia and North Carolina. Final Report to MARFIN, National Oceanic and
Atmospheric Administration. 141 pp.
Blount, Ben G., and Kathi R. Kitner. 2007. Life on the Water: A Historical-Cultural Model of
African American Fishermen on the Georgia Coast (USA). NAPA Bulletin (National
Association of Practicing Anthropologists). 28(1): 109-122.
Boyd, Heather, and Anthony Charles. 2006. Creating Community-based Indicators to Monitor
the Sustainability of Local Fisheries. Ocean and Coastal Management. 49(2): 237-258.
Buck. E. H., 2007. Fishery, aquaculture, and marine mammal legislation in the 109th Congress.
Congressional Research Service. CRS Report to Congress. RL33459. Congressional
Research Service, Washington, DC
Buckle, Philip, Graham Marsh, and Sydney Smale. 2001. Assessing Resilience and
Vulnerability: Principles, Strategies, and Actions. Guidelines Prepared for Emergency
Management Australia. Victoria, Australia: Emergency Management Australia.
Bureau of Labor Statistics (BLS). 2002. Bureau of Labor Statistics Data: Public Data Query.
U.S. Dept. of Labor, Washington, DC. <http://data.bls.gov/cgi-bin/surveymost>
Chang, T.C. 2000. Theming Cities, Taming Places: Insights from Singapore. Geograkiska
Annaler. Series B, Human Geography. 82(1): 35-54.
Clay, Patricia M., and Julia Olson. 2008. Defining "Fishing Communities": Vulnerability and the
Magnuson-Stevens Fishery Conservation and Management Act. Human Ecology Review.
15(2): 143-160.
Cutter, Susan L., Jerry T. Mitchell, and Michael S. Scott. 2000. Revealing the Vulnerability of
People and Places: A Case Study of Georgetown County, South Carolina. Annals of the
Association of American Geographers. 90(4): 713-737.
Cutter, Susan L. (2003). GI Science, Disasters, and Emergency Management. Transactions in
GIS, 7(4):439-445.
Cutter, Susan L. Byron J. Boruff, and W. Lynn Shirley. (2003). Social Vulnerability to
Environmental Hazards. Social Science Quarterly, 84(2):242-261.
97
DeSantis, J. 2008. Shrimp make a showing, but size limits marketing. National Fisherman. 89,
August: 12.
Donoghue, Ellen M., and Victoria E. Sturtevant. 2007. Social Science Constructs in Ecosystem
Assessments: Revisiting Community Capacity and Community Resiliency. Society and
Natural Resources. 20(4): 899-912.
Dow, Kirstin. 1999. The Extraordinary and the Everyday in Explanations of Vulnerability to an
Oil Spill. Geographical Review. 89(1): 74-93.
Federal Register. 1998.
Washington D.C.
632(84): 24235.
May 1, 1998.
Government Printing Office.
Fesenmaier, D. R., S. Um, W.S. Roehl, A. S. Mills, T. Ozuna, Jr., L.L. Jones, and R. Guajardo.
1987. Trinity-San Jacinto Estuary: Economic Impact of Recreational Activity and
Commercial Fishing. A Report to the Texas Water Development Board, Texas
Agricultural Experiment Station, College Station.
Flint, Courtney G., and A.E. Luloff. 2005. Natural Resource-Based Communities, Risk, and
Disaster: An Intersection of Theories. Society and Natural Resources. 18(5): 399-412.
Frere, Julian, and Pierre Failler. 2001. Regional Socio-Economic Studies on Employment and the
Level of Dependency on Fishing in England and Wales. Salerno, Italy, Paper presented
at the 13th Annual Conference of the European Association of Fisheries Economists.
Freudenberg, William R. 1992. Addictive Economies: Extractive Industries and Vulnerable
Localities in a Changing World Economy. Rural Sociology. 57(3): 305-332.
Freudenburg, William R., and Robert Gramling. 1992. Community Impacts of Technological
Change: Toward a Longitudinal Perspective. Social Forces. 70(4): 937-955.
Gallaway, Alecya.. 2002. The Human Role: Past. In J. Lester and L. Gonzalez, Eds. The State
of the Bay: A characterization of the Galveston Bay ecosystem. Pp 24-38. Houston, TX:
Galveston Bay Estuary Program, GBEP T-7, AS-186/02
Gottdiener, Mark D., and Leslie Budd. 2005. Key Concepts in Urban Studies. Thousand Oaks,
CA: Sage Publications, LTD.
Hawley, Amos H. 1986. Human Ecology: A Theoretical Essay. Chicago: The University of
Chicago Press.
Hamnett, Chris. 1991. The Blind Men and the Elephant: The Explanation of Gentrification.
Transactions of the Institute of British Geographers. 16(2): 173-189.
98
Hartley, Troy W., Michele Gagne, and Robert A. Robertson. 2008. Cases of Collaboration in
New England Coastal Communities: An Approach to Manage Change. Human Ecology.
15(2): 213-226.
Impact Assessment, Inc. 2004. Identifying Communities Associated with the Fishing Industry in
Louisiana. A report prepared for U. S. Department of Commerce, NOAA Fisheries,
Southeast Region St. Petersburg, Florida
Impact Assessment, Inc. 2005a. Identifying Communities Associated with the Fishing Industry
along the Florida Gulf Coast. U. S. Department Of Commerce NOAA Fisheries,
Southeast Regional Office St. Petersburg, Florida. Contract WC133F-02-SE-0298.
Impact Assessment, Inc. 2005b. Identifying communities associated with the fishing industry in
Texas. A report prepared for U. S. Department of Commerce, NOAA Fisheries,
Southeast Region St. Petersburg, Florida Contract WC133F-03-SE-0603. 424 pp.
Impact Assessment, Inc. 2006a. Preliminary Assessment of the Impacts of Hurricane Katrina on
Gulf of Mexico Coastal Fishing Communities. Final Technical Report submitted to U. S.
Department Of Commerce NOAA Fisheries, Southeast Regional Office St. Petersburg,
Florida. Contract # WC133F-06-CN-0003
Impact Assessment, Inc. 2006b. Identifying Communities Associated with the Fishing Industry
in Alabama and Mississippi. U. S. Department Of Commerce NOAA Fisheries,
Southeast Regional Office St. Petersburg, Florida. Contract WC133F-03-SE-0603.
Jacob, Steve, Lisa Bourke, and A.E. Luloff. 1997. Rural Community Stress, Distress, and WellBeing: A Pennsylvania Assessment. Journal of Rural Studies. 13(3): 275-2888.
Jacob, Steve, Frank L. Farmer, Jepson, Michael, and Charles Adams. 2001. Landing a Definition
of Fishing Dependent Communities: Potential Social Science Contributions to Meeting
National Standard 8. Fisheries. 26(10): 16-22.
Jacob, S. and M. Jepson. 2009. Creating a Community Context for the Fishery Stock
Sustainability Index. Fisheries. 43(5): 228-231.
Jacob, Steve, Michael Jepson, and Frank L. Farmer. 2005a. What you see is not always what you
get: Aspect dominance as a confounding factor in the determination of fishing dependent
communities. Human Organization. 64(4): 373-384.
Jacob, Steve, Michael Jepson, Carleton Pomeroy, David Mulkey, Charles Adams, and Suzanna
Smith. 2002. Identifying Fishing Dependent Communities: Development and
Confirmation of a Protocol. Marine Fisheries (MARFIN) Project Report NA87FF0433,
St. Petersburg, NOAA NMFS.
99
Jacob, Steve, A.E. Luloff, and Jeffrey C. Bridger. 2005b. Pennsylvania Rural Communities and
Individual Mental Health. Journal of Rural Community Psychology. VE8(1): 1-24.
Jacob, Steve, and Fern K. Willits. 1994. Objective and Subjective Indicators of Community
Well-Being: a Pennsylvania Assessment. Social Indicators Research. 32(2): 161-171.
Jepson, Michael Edward. 2004. The impact of tourism on a natural resource community: Cultural
resistance in Cortez, Florida. Ph.D. in Applied Anthropology diss., University of Florida.
Jepson, Michael, and Steve Jacob. 2007. Social Indicators and Measurements of Vulnerability
for Gulf Coast Fishing Communities. NAPA Bulletin. 28: 57-68.
Jepson, Michael, Kathi Kitner, Ariana Pitchon and W. Perry. 2005. Fishing Communities in the
Carolinas, Georgia and Florida: An Effort at Baseline Profiling and Mapping. South
Atlantic Fishery Management Council, Charleston, SC.
Khan, Nuzrat Yar. 2007. Multiple Stressors and Ecosystem-Based Management in the Gulf.
Aquatic Ecosystem Health and Management. 10(3): 259-267.
Landis, J.R., and G. G. Koch. 1977. The measurement of observer agreement for categorical
data. Biometrics. 33:159-174.
Lees, Loretta. 2000. A Reappraisal of Gentrification: Towards a 'Geography of Gentrification'.
Progress in Human Geography. 24(3): 389-408.
Lewis, Glenn. 1979. Troubled Waters: Vietnamese, Texans vying sharply for shrimp business.
Houston Post. Nov. 11. Section A, Page 1, Column 2.
Lombard, M., J. Snyder-Duch, and C. C. Bracken. 2002. Content analysis in mass
communication: Assessment and reporting of intercoder reliability. Human
Communication Research. 28:587-604.
Luloff, A.E., and L. Swanson. 1995. Community Agency and Disaffection: Enhancing Collective
Resources. In Investing in People: The Human Capital Needs of Rural America,
Beaulieu, L; Mulkey, D (eds). 351-372. Boulder, CO: Westview Press.
Maine Sea Grant, Marine Extension Team. 2007. Access to the Waterfront: Issues and Solutions
Across the Nation. MSG-E-07-04, Orono, ME, Maine Sea Grant.
Manyena, Siambabala Bernard. 2006. The Concept of Resilience Revisited. Disasters. 30(4):
433-450.
McEntire, David A., Christopher Fuller, Chad W. Johnston, and Richard Weber. 2002. A
Comparison of Disaster Paradigms: The Search for a Holistic Policy Guide. Public
Administration Review. 62(3): 267-281.
100
Molotch, Harvey. 1976. The city as a growth machine: Towards a political economy of place.
American Journal of Sociology. 82(1): 309-332.
Oliver-Smith, Anthony. 1996. Anthropological Research on Hazards and Disasters. Annual
Review of Anthropology. 25: 303-328.
NOAA Fisheries Service (NMFS). 2008. Fish Stock Sustainability Index, Status of U.S.
Fisheries. NOAA Fisheries Service, Office of Sustainable Fisheries, Silver Spring,
Maryland. Available at: Http://www.nmfs.noaa.gov/ sfa/domes_fish/ StatusoFisheries/
2008/1stQuarter/Q1-2008-FSSI SummaryChanges.pdf.
Parkins, John R., Richard C. Stedman, and Thomas M. Beckley. 2003. Forest Sector Dependence
and Community Well-Being: A Structural Equation Model for New Brunswick and
British Columbia. Rural Sociology. 68(4): 554-572.
Peluso, Nancy L., Craig R. Humphrey, and Louise P. Fortmann. 1994. The Rock, The Beach,
and the Tide Pool: People and Poverty in Natural Resource Dependent Areas. Society and
Natural Resources. 7(1): 23-39.
Perkins, Daniel F., and Judy R. Butterfield. 1999. Building an Asset-Based Program for 4-H.
Journal of Extension (<http://joe.org/joe/1999august/a2.html>) 37, 4.
Phillips, Martin. 2002. The Production, Symbolization and Socialization of Gentrification:
Impressions from Two Berkshire Villages. Transactions of the Institute of British
Geographers. 27(2): 282-308.
Richardson, Harry W. 1979. Regional Economics. Urbana, IL: University of Illinois Press.
Rossi, Robert J., and Kevin J. Gilmartin. 1980. The Handbook of Social Indicators: Sources,
Characteristics, and Analysis. New York: Garland STPM Press.
Sheridan, Francis. 1954 [in Gallaway 2002]. Galveston Island or a Few Months off the Coast of
Texas., The Journal of Francis Sheridan 1839-1840. Willis W. Pratt, editor. University of
Texas Press. Austin, Texas.
Smith, Suzanna D., Steve Jacob, and Michael Jepson. 2003. The Stress Process in Florida's
Commercial Fishing Families. Society and Natural Resources. 16(1): 39-59.
Smith, Suzanna, and Michael Jepson. 1993. Big Fish, Little Fish: Politics and Power in the
regulation of Florida's Marine Resources. Social Problems. 40(1): 39-49.
St. Martin, Kevin, and Madeleine Hall-Arber. 2008. Creating a Place for "Community" in New
England Fisheries. Human Ecology Review. 15(2): 161-170.
101
Stedman, Richard C., John R. Parkins, and Thomas M. Beckley. 2004. Resource Dependence
and Community Well-Being in Rural Canada. Rural Sociology. 69(2): 213-234.
Texas Coastal and Marine Council. 1983. Texas Bay Shrimp Industry. Status Report and
Recommendations. Executive Summary.
Tuler, Seth, Julian Agyeman, Patricia Pinto de Silva, Karen Roth LoRusso, and Rebecca Kay.
2008. Assessing Vulnerabilities: Integrating Information about Driving Forces that Affect
Risks and Resilience in Fishing Communities. Human Ecology. 15(2): 171-184.
Turner II, B.L., Roger E. Kasperson, Pamela A. Matson, James J. McCarthy, Rober W. Corell,
Lindsey Christensen, Noelle Eckley, Jeanne X. Kasperson, Amy Luers, Marybeth L.
Martello, Colin Polsky, Alexander Pulsipher, and Andrew Schiller. 2003. A Framework
for Vulnerability Analysis in Sustainability Science. Proceedings of the National
Academy of Sciences. 100(14): 8074-8079.
Weeks, Priscilla, and Jane M. Packard. 1997. Acceptance of Scientific Management by Natural
Resource Dependent Communities. Conservation Biology. 11(1): 236-245.
Wildin, Brooke, and John Minnery. 2005. Understanding City Fringe Gentrification: The Role of
a 'Potential Investment Gap'., Brisbane, AU, Paper Presented at the State of Australian
Cities 2005 National Conference.
Wilkinson, Kenneth P. 1999. The Community in Rural America. Middleton, WI: Social Ecology
Press.
Yagley, James, Lance George, Cequyna Moore, and Jennifer Pinder. 2005. They Paved
Paradise…Gentrification in Rural Communities. Report Prepared for U.S. Department of
Housing and Urban Development (HUD), Washington, DC, Housing Assistance Council.
Zukin, Sharon. 1995. The Culture of Cities. Oxford: Wiley-Blackwell.
102
Appendix A
Semi-Structured Interview Protocol
103
INTERVIEW SUMMARY SHEET
ID:
DATE:
COMMUNITY:
LICENSE TYPE:
ETHNICITY:
AGE (YRS): 1) 19-24
2) 25-29
3) 30-34
4) 35-39
5) 40-44
6) 45-49
7) 50-54
8) 55-60
2) 1000 – 2500
3) 2600 - 4000
9) OVER 60
MONTHLY INCOME LEVEL ($):
1) LESS THAN 1000
SUMMARY ANSWERS FOR EACH QUESTION (BULLETS):
1. How long have you fished in this area?
2. Did your parents and grandparents fish also?
3. Is fishing the biggest source of income for your family?
4. Currently, whom do you sell your catch to?
5. In your opinion, how much has fishing in this area changed?
6. How has the fishery changed?
104
7. What do you think has caused changes in fishing as a way of life?
8. What do you think has caused changes in the fishery?
9. Would you say that you live in a fishing community?
10. How would you say you’re doing?
11. How have you adapted to fishing changes in the past?
12. Do you belong to any fishing or other support group?
13. What appeals to you about fishing?
14. What might make you want to leave the fishery?
15. What do you say is the biggest problem for you as a fisherman?
What is the biggest problem for your community?
16. Miscellaneous:
105
Appendix B
Historical Sketches of the Coastal Communities
106
Galveston Bay Communities
The shores of Galveston Bay have been inhabited for thousands of years, and the
existence of large shell middens points to the first inhabitants being fishermen. Later Native
American groups include the Karankawa and Tonkawa, who also relied heavily on various forms
of shellfish, and most likely, finfish. From prehistoric times to the present, fishing has been a
major human activity in the Bay.
Early Anglo settlers were attracted to the region by the herds of Longhorn cattle,
descendents of Mexican and Spanish cows brought to Texas by early missionaries in the 1690s.
By the 1800s, the cattle had become native to the area and could be considered wild. Cowboys
from Louisiana came to Texas and taught others to manage the marshes for cattle production.
Eventually seven “breeds” of Longhorns came from this region (Gallaway, 2002). There was
environmental interplay, as cowboys used to run the cows across an oyster bar (which was
dredged for shell during WWII and no longer exists) to the other side of the bay, and they also
rustled cows using barges. Cattle were a key resource in the 1800s, and there were tanneries and
slaughterhouses in Dickinson and San Leon.
Italian and Japanese immigrants also produced rice, fruit and cotton in the 1800s. At one
point, Dickinson was known as the strawberry capital of Texas. The Italian migration to
Galveston Bay was promoted by the then Italian Consul in Galveston who owned land in the
region.
Figure 4: Cattle Drive across Redfish Bar, Galveston Bay. Courtesy Wallisville (TX) Museum.
107
San Leon is the oldest fishing community on Galveston Bay, dating back to 1838. In the
1830s, fishermen sold a variety of species including oysters, flounder, redfish, skate and shrimp,
the latter of which were collected by seine (Sheridan, 1954 [in Gallaway, 2002]). Oysters were
limited to the local trade until the late 1800s when new processing and shipping technologies
were developed. Oyster production increased to the point of resource depletion which was the
catalyst for the creation of the Oyster Commission, later to become the Texas Parks and Wildlife
Department. The Commission encouraged oyster growing on private leases, instituted in 1912,
which are still an important component of the fishery today. The removal of the mud shell from
the bottom of the bay during WWII negatively impacted oyster production, and post-WWII,
oyster shell became a resource for concrete and for roads, creating conflict between oyster
fisheries and the industries that used the shell. In San Leon, the mud shell industry devastated
the fishing industries in late 50s and early 60s. Shell dredging was banned in the mid-to-late 60s,
but by that time so much of the shell and hard substrate had already been removed, which
changed the ecological dynamics of fishing. Not only were oyster fisheries impacted, but also
those fisheries that depended on submerged grasses which were silted over due to constant shelldredging.
By the late 1800s, the railroad connected the bay communities to the growing cities to the
north, not only expanding commercial fishing but making the communities important coastal
resort areas. Clear Lake Shores offered building lots for opening a checking account. The city
of Seabrook was developed by the Clear Creek Development Company as a resort community
and was named after one of the members of that company. Investors proposed San Leon as the
next Atlantic City. Recreational boating became popular, and both Kemah and Seabrook had
boatyards to build recreational and travel boats.
In the 1920s, shrimp became an important fishery in Galveston Bay although shrimp was
not popular locally. Shrimp was exported to Japan, and with the introduction of gas motors and
canneries, shrimp eventually became the leading fishery. Shrimp became a popular food with
locals on the return of GIs from Asia in WWII, to the extent that shrimp became a symbol of the
region. In the city of Galveston, the traditional Oleander Bowl was re-named the Shrimp Bowl
in the 1950s. At the same time, the channel between Clear Lake and Galveston Bay was opened,
and Kemah and Seabrook became home to a large shrimp fleet, promoted by Italians that
migrated north from Galveston Island. Canneries and docks dotted the shoreline.
108
Figure 5: Shrimp Ready to be Packed in Barrels on the Galveston Dock, 1920s.
In the 1900s, major ports developed in Texas City and Galveston, and agricultural and
manufactured goods were shipped from these ports. The deepwater port at Texas City
contributed to its development as an industrial center. The Texas City Refining Company was
created in 1908, precipitating a building spree of industrial infrastructure. Texas City soon
became home to several refineries, tank farms, and its landscape was crisscrossed with pipelines.
It is now an important center of the processing and chemical producing industries on the Gulf
Coast as well as a port which handles these commodities.
After the 1900 hurricane devastated the city of Galveston, the city of Houston deepened
and widened Buffalo Bayou to Galveston Bay to provide access to large ocean-going ships. This
paved the way for increased industrialization in the communities on the western shore of the bay
and petrochemical plants are concentrated in several bayside communities.
Modern
suburbanization began when NASA’s Johnson Space Center was built in 1962 on the West
Ranch. Other parts of the ranch were taken for housing developments. The University of
Houston built a satellite campus in the early 1970s and large engineering firms associated with
NASA and the petrochemical industries built offices to be close to NASA. Population grew
rapidly during the last three decades of the twentieth century, and the Houston and Galveston
Bay area became one of the major population centers in the United States. New suburbs are still
appearing, and others are in the planning stages. Communities like Seadrift, Kemah, Bacliff,
and San Leon become compacted within the larger urban area, and boundaries among them
become indistinct.
Tourism and travel are major producers of income in the Houston/Galveston Bay region.
Total travel spending in 2001 for the Galveston Bay area was about $6.2 billion, representing
109
70.6 percent of all travel spending for the Texas coast and 21.3 percent of all travel spending for
the entire state. Most travel spending is related to business and non-water related recreation, but
the latter is economically important. The 1986 Fesenmaier Survey estimated bay/estuary related
recreational expenditures at $294 million (Fesenmaier et al., 1987). That represented 11.5
percent of the estimated $2.56 billion of total travel expenditures in the Galveston Bay area in
1986. This proportion is applied to all years subsequent to 1986 because no other estimate of
bay/estuary related recreational expenditures exists for the area. The estimate for bay/estuary
spending in 2001 is $708 million. When adjusted for inflation using the CPI for the HoustonGalveston-Brazoria area (BLS, 2002), real growth (in constant 2001 dollars) of Galveston B&E
related spending grew by about 41 percent from 1992-2001.
San Antonio, Lavaca, Matagorda, Bays Communities
Like Galveston Bay, the lower coast of Texas was originally inhabited by Native
American communities, all heavily dependent on fishing. After the decline of the Karankawa
tribes, the major inhabitants of the shorelines, other groups moved into the region, first the
Tejanos (Texans of Mexican descent living within the territory that Texas took from Mexico)
who had land grants, and subsequently, German immigrants. Small farmers were lured to the
region by low interest loans. By the 1930s, agribusiness began aggregating lands to develop
commercial farms. In contrast to the Galveston area, 70% of this region was still in farms and
ranches in the 1980s.
Port Lavaca, the largest city and the county seat in Calhoun County, was originally
named La Vaca (the cow). It was established as a port and soon became a major shipping center,
providing transit for cotton, hides, tallow, horns, an array of ores, wool, pecans and wine.
During the 1920s and 30s, Port Lavaca became one of the leading ports for the shipping of
shrimp and oysters. It did, however, always have a diversified economy that included
agriculture, manufacturing, mineral extraction, tourism, commercial hunting and fishing as well
as shipping. The town boasted a large pavilion with a water slide into the bay for tourists.
Figure 6: Early Twentieth Century Shoreline of Port Lavaca, Texas.
110
Greek immigrants played a large role in the fishery in Calhoun County in the early 1900s.
At that time Lavaca and Matagorda Bays had substantial oyster shell reefs. Sailing skiffs were
used to harvest oysters. Businessmen, with non-fishing holdings, saw the potential of seafood in
the region and began building a fishery infrastructure that included fishing fleets as large as 70
vessels, fish houses and processing plants. By the 1920s, Port Lavaca was a leading shipper of
seafood in the U.S. Rail became an important method of shipping, and the process of de-heading
shrimp and icing them in barrels for train shipment was begun by a Port Lavaca businessman.
There was also a municipal dock and freezing plant.
Seadrift was also historically a port town for both fishing and shipping. Exports included
seafood and vegetables. In the early 1900s, it had a canning factory, an ice factory, a cotton gin
and four hotels. By 1914, it had 1,250 residents and was served by passenger train. A hurricane
caused the town to contract to about a third of that size, but it gradually built up again and was
slightly less than 1,000 in the 1970s when Vietnamese refugees resettled there.
Like the city of Seabrook, the towns of Port O'Connor and Palacios were established by
developers, Port O’Connor as a summer resort area, and Palacios for settlement. Both were
formerly part of large cattle ranches that were sold to development firms, and both were
promoted by railroads as excursion spots. The latter led to the construction of infrastructure such
as bath houses, piers, hotels and dancing pavilions to accommodate tourists. Today, Port
O’Connor remains a resort area, but Palacios still has a large fishing fleet and accompanying
infrastructure, including public docks. Palacios has only recently begun to cater to tourists once
again. Neither town is home to heavy industry but both are close enough to towns with industry
for its citizens to live in Palacios and commute.
Similar to Texas City, but on a smaller scale, Port Lavaca’s port served as the catalyst for
the growth of heavy industries which arrived in the early 1940s. Unlike Galveston Bay with
several large concentrations of industrial parks, industrial facilities are scattered throughout the
region.
The In-migration of Vietnamese Fishermen to Texas
Both research regions received Vietnamese immigrants during the 1970s and 1980s, and
conflict in both regions was severe. In some instances, it divided the Anglo fishing community.
Dock owners and fish houses that worked with the Vietnamese were criticized by fishermen who
felt their livelihood was being threatened.
Approximately 250,000 Vietnamese came to Texas, 23-25,000 to the Houston-Galveston
region. Although official federal policy was to disperse the refugees throughout the nation in
order to avoid enclaves, many immigrants migrated from their community of original settlement
to communities with friends or relatives. Several of our interviewees came to Texas after being
111
settled in the mid-western U.S. and entered fishing through relatives. Some were fishermen in
Vietnam, but many others were not.
Vietnamese fishermen participated in the shrimp and crab fisheries. They were accused
by resident fishermen of overharvesting the resource, not registering boats, using illegal gear,
and violating norms of fishing, including, for crabs, informal territories. Conflict in our research
regions was severe during the late 1970s and early 1980s. In the Galveston region, there were
two Klan rallies complete with cross and boat burnings, a civil lawsuit by the Vietnamese against
the Klan and its fishermen supporters, and shotgun-wielding Klan members on ‘parade’ on
shrimp boats. In Seadrift, two Vietnamese fishermen killed an Anglo fisherman who they
claimed was threatening them. The crab plant where many Vietnamese worked was forced to
close temporarily. About a third of its workers were Vietnamese, and all but two of about 25
Vietnamese families moved away due to fear of reprisals.
To ease tensions in the shrimp fishery, and to give them time to investigate the impact of
the influx of newcomers on the fishery, the State Legislature, which at that time regulated the
fishery directly, called for a two year moratorium on new shrimp licensees.
The impact of the influx of migrants into shrimp and crab fisheries, however, is not clear.
A 1979 study for the Commerce Department by Trans-Century Corporation (Lewis, 1979),
concluded that the shrimp fishery was stretched to the limit and that the influx of newcomers was
partially responsible. It recommended the diversification of fisheries and suggested the creation
of a soft shell crab industry and aquaculture to lessen pressure on the shrimp fishery. A contrary
view was offered by the Texas Coastal Marine Council (1983), which was directed by the Texas
Legislature and which at that time managed the shrimp fishery directly, to assess the status of the
shrimp fishery in Texas. It concluded that the impact of the Vietnamese fishermen were one of
many stresses facing the fishery, noting the 800% increase in fuel. Furthermore, it concluded
that biological overfishing was not a problem and recommended the expiration of the
moratorium. A separate inspection of all boats in the fishery for illegal rigging and inadequate
documentation found no significant differences in compliance between non-Vietnamese and
Vietnamese shrimpers.
After several decades of Vietnamese presence in the Galveston, Matagorda, and San
Antonio Bays, open conflict rarely, if ever, occurs. Lingering bad feelings, however, still exist
within these communities, as evidenced by some of the comments from our interviewees about
overcrowding and Vietnamese.
112
Appendix C
Gentrification: Communities
113
Gentrification
Mainland Galveston Bay Communities
The communities on the western shore of Galveston Bay have undergone an array of land
use changes since they were first platted as agricultural lands. The creation of NASA in the late
1960s, followed by the siting of a satellite campus for the University of Houston, began a
housing boom that has increased steadily ever since, transforming prairies and fields into homes.
This transformation has occurred in a patchwork fashion and has resulted in uneven
gentrification among these communities.
San Leon-Bacliff
San Leon and Bacliff, referred to collectively as the bayshore communities, have thus far
not gentrified vis-à-vis sub-developments and recreational marina projects. San Leon could be
said to be the last “fishing community” along Galveston Bay, although commercial fishers in that
town tended to note that the community is more a past relic than a present day reality. San Leon
historically has had a more pronounced fishing community, while Bacliff developed more along
the lines of a second-home/vacation community. Bacliff and San Leon remain unincorporated
and non-zoned, with two subdivisions in Bacliff and one gated subdivision in San Leon. A
review of census data shows that median household incomes in these two places are the lowest
amongst the Galveston Bay communities surveyed and also contain the oldest housing stock. In
2000, certain residents of both communities lobbied publicly for the incorporation of the two
places into what would have been the village of Baytown. However, that initiative failed to pass
a public referendum. In 2005, there was a brief public discussion of annexation of San Leon by
Texas City and of Bacliff by League City which respectively have extra-territorial jurisdiction
over these communities. At this point, the cost-benefit calculus does not appear favorable for
either municipality to develop these bayshore communities.
Kemah-Seabrook
Of the several communities surveyed, Kemah has displayed the most pervasive
conversion of its waterfront from commercial fishing to recreational development, a change that
is perhaps most strikingly epitomized by the growth of the Kemah Boardwalk in the late 90s and
the recreational boating marina nearby. It is also striking that at the Kemah City Hall, the
council chambers are surrounded by photographs of local shrimp boats from the 1960s and 70s,
while the contemporary seal of the city includes only a recreational sailboat. Hurricane Carla
and land subsidence threatened the bay front areas of Kemah during the 1960s and 70s. Carla
destroyed 60% of businesses and 30% of the housing stock in what was then an unincorporated
village. At its most acute point in time, some parts of Kemah were sinking as much as 12
inches-per-year. In addition to the expected economic vitalization, proponents of the marina
argued that its development would be the only way to claim what would otherwise have been
commercially worthless property along the waterfront. The dredged spoils from the marina
114
would then be used to elevate the surrounding waterfront businesses. The city of Kemah had
pursued marina projects since its municipal incorporation in 1965. The city was able to facilitate
the necessary land acquisitions through the Urban Renewal Agency Act in the early 1980s. In
1983, the citizens of Kemah voted by a 4-1 margin in favor of marina development. Clear Lake,
once home to a large bait shrimp fishery, now hosts the 3rd largest recreational fleet in the
country.
The seafood restaurants on the Boardwalk are part of a national restaurant chain and rely
virtually exclusively on imported seafood rather than local catches. Currently shrimpers dock
their boats underneath the bridge between Clear Lake and Galveston Bay at no charge but this is
on the condition that they sell their catch to the owner of the dock located on the Seabrook-side
of the channel. According to one of the shrimpers who docks there, the owner is looking to sell
the docks because the city of Kemah has eyed this area as a potential site of capital infusion.
Redevelopment of the boat ramp may ultimately displace the remaining commercial shrimpers.
The city of Kemah has given consideration to partnering with Galveston County to
submit an application to the State of Texas Boat Ramp Program to develop the ramp underneath
the bridge. This could be a catalyst for luring more recreational tournaments for Kemah.
Currently, the boat ramp and three docks are within Kemah city limits (County-owned), while
the southern part where the shrimpers dock, is within the limits of Clear Lake Shores. Plans for
the gentrification of the Clear Lake Shores dock have been folded into a broader plan to increase
eco-tourist zones within that city.
One of the most potent symbols of the gentrification of the Kemah-Seabrook channel is
the refashioning of the traditional Blessing of the Fleet, begun by Italian fishermen. It is now
promoted by chambers of commerce, and the public views the blessing from the Kemah
Boardwalk. It is replete with a Mrs. Blessing of the Fleet, a Master of Ceremonies, a Protestant
as well as a Catholic priest to bless the fleet, yachts, some fishing boats, dinghies etc. and a prize
for best decoration. What was once an ethnic religious ceremony associated with livelihood,
safety and work has been recreated sans ethnicity and religion with little affiliation to work.
Kemah recently annexed new lands and three projects are planned (See Appendix D for
list). Entrepreneurs are buying property in Kemah and Seabrook in the hopes that gambling will
be legalized in Texas. Various gambling bills have been put before the Legislature over the
years including a cruise ship terminal in the area, on-land casinos or a gambling easement.
The City of Galveston
Galveston is a key tourism destination for Houstonians as well as a site for second homes.
Tourism is currently the fastest growing industry for Galveston Island (AngelouEconomics,
2008). There are two primary tourism foci – nature based tourism, which includes beaches, bird
trails and recreational fishing, and historical tourism. Other important tourist attractions include
Moody Gardens, Mardi Gras, and the Lone Star Motorcycle Rally.
115
The Galveston Island Nature Based Tourism Council promotes a variety of nature based
tourism activities including an annual birding festival, nature education signage on the beach, a
web based guide to birding spots and the development of a master plan for the city’s nature
preserve. Beaches bring the majority of tourists to Galveston each year and significant funding
is dedicated to beach replenishment. Recreational fishing is also an important component of
tourism and is slowly squeezing out commercial fisheries. There are charter operations on the
Harborside further east near the Bob Smith Yacht club and basin next to it. This is the likely
area for intensive development and investment for residential and tourist industries. Currently,
the ‘Mosquito Fleet’ and other fishing boats dock at the public harbor nearby. The harbor is
owned by the city and leased to fishermen. This lease is set to expire soon and it is not yet
certain whether it will be renewed or not.
One catalyst for gentrification in Galveston has been The Galveston Historical
Foundation, which was incorporated in 1954. Projects include the revitalization of the historic
downtown area, including The Strand and several streets adjacent to it, the restoration of an 1877
tall ship named the Elissa, the creation of historic districts, revitalization of historic
neighborhoods and the old Opera House and historic festivals and home tours.
Additionally, some of the old piers and warehouses have been converted to hotels and
restaurants. There has been talk of converting one of the piers into a Kemah Boardwalk type of
development.
The recently revitalized downtown area is close to the University of Texas Medical
Branch (UTMB), Texas’ oldest medical school and the city of Galveston’s major employer for
middle class and professional residents. The residential areas surrounding UTMB are a mix of
revitalized historic homes and public housing. A coalition of downtown business owners is
promoting a ‘walkable cities’ approach to downtown Galveston with walking streets in which
cars are prohibited.
In addition to revitalizing already built neighborhoods, the City is encouraging new
development. It has instituted several Tax Reinvestment Zones in which it reimburses
developers for investment in infrastructure. A study of Galveston’s economy commissioned by
one of the developers of East End Flats indicates that Galveston has not had significant nonvacation type homes built in 35 years and consequently middle to upper middle class home
owners have moved to newer housing on the mainland. Over half of Galveston’s employees live
on the mainland.
There are several large developments being created on the beach. Palisade Palms is a
high rise condominium built to state of the art hurricane standards not far from downtown. The
themed community, Beachtown, is located next to the Palms on the beachfront. Both have
received public funds. New development is also planned in the East End flats a former
government installation. Before the hurricane (Appendix E), the City and UTMB were
116
promoting the development of an initial 93 acres of the flats. Further from downtown, on west
beach, Marquette Development is proposing a master planned community of no more that 3,948
living units. It will be a mixed use development with a resort hotel, nature preserve and golf
course. This development in particular is unpopular with conservationists because of its
sensitive location.
San Antonio and Matagorda Bays
Coastal land in Calhoun County consists of large tracts of privately held lands that have
begun to be sold to developers in the last few years. These new developments include The
Sanctuary and Caracol. The Sanctuary is being built on 12,000 acres, with financing from
investors from North Carolina, South Carolina, and Florida. There are conflicting reports of
how much the lots are going for. A retired teacher who was interviewed claimed that lots are
going for $149,000. According to a representative from the Calhoun County EDC, a roughly
60’x180’ lot had recently increased in price from $200,000 to $400,000. This representative
stated that approximately 800 have been sold. According to one of the project developers
interviewed, of the 12,000 acres purchased, only 1,100 have been developed, and there are no
plans to develop the rest. The Sanctuary is marketing itself as ecologically friendly and new
wetland acreage is being created. According to the developer, only 550 lots have been sold or
are under contract. All lots are either on the water or along the road. Although the clubhouse,
marina post office boxes, BBQ pavilions, and fish cleaning stations are all almost nearly
complete and the utility infrastructure has been installed, no houses had yet been built at the time
of the field work. The development is positioned approximately 300-600 yards from the
Intracoastal and 8 miles from the Gulf. All home sites that are by the water have boat access and
individual jetties for groups of home sites.
The other major new residential development is “Caracol.” The asking price for a lot can
be as high as $600,000, according to a representative from the Calhoun County Economic
Development Corporation who stated that a corner lot had recently sold for that amount.
One interview with a man who runs a bait house, restaurant and hotel—and until recently,
had his son running fish house—exemplified the effects of gentrification on local fishing. The
ranch across the road from where his property is located allegedly pays $1,900 a year in property
taxes for 13,000 acres, “because they grow cattle on it.” This man paid $86,000 in property
taxes last year and expects a similar bill this year. He had to borrow money from the bank to pay
his taxes. He paid $50,000 on the property alone since the rate is $3,000 per waterfront foot.
According to him, “your land is as valuable as the last sale…so that if someone pays $500,000
for a lot, your property is considered just as valuable.”
Port O’Connor has a full-time residential population of approximately 2,200. One
interviewee claimed that about 15% of the daily population is full-time residential, while a local
real estate agent put that figure at 25%. Port O’Connor was originally planned as a residential
117
and vacation community prior to the emergence of commercial fishing in the 1930s. Following
Hurricane Carla, the development of vacation properties escalated. By 1985, the five shrimp
wholesalers were out of business. Informants generally seem to agree that anywhere from 5-10
years ago, there was another intensification of new recreational development, although none of
the interviewees could definitively state why. A Municipal Utility District was established 6
years ago. Besides The Sanctuary and Caracol, other recent developments include Baypoint and
the Powderhorn Ranch (which used to be a lodge for recreational fishers) half-acre lots on its
8,500 acres are selling for $350,000. According to a representative from the EDC, there have
been rumors that a golf course may be installed in the future.
Palacios
According to interviewees, shrimping groups in Palacios have remained active longer
than in Port Lavaca or Port O’Connor, although the industry’s decline in recent years has
paralleled the Navigation District’s development of a “more diverse phase of economic
development.” The District owns over 300 acres of near shoreline property within a Foreign
Trade Zone. Unlike Port O’Connor, the county is the largest landowner in Palacios. The largest
new residential development is the Beachside project that overlooks Turtle Bay. It is located on
the former Camp Hulen, which was sold forty years ago to a private company, and then resold
two years ago to Cherokee Development. According to a local banker who was interviewed, the
corporation has put $20m into the project, including a $10m deposit for the installation of roads,
water, and sewer. The development is being built in four phases with a total of 3,000 lots
expected to be offered. They have sold 300 thus far. Another interviewee stated that Beachside
was the only new development, but there would likely be more with construction of a nuclear
power plant. A hardware retailer who was interviewed said that a rumor is circulating that
property in Palacios would be reappraised once the first home goes up in Beachside. He stated,
albeit vaguely, that taxes had already gone up on account of Beachside and feared that
reappraisals would “run people out of town who can’t afford the higher taxes.”
One interviewee who was in “harbor management,” stated that, “commercial boats are
being pushed out for residential and resort development all over this coast. That’s our bread and
butter. That’s a good opportunity for Palacios.” Further, the general manager of a specialty boat
building & metal cutting company conceded that, “everyone has visions that Palacios will turn
into Port O’Connor” but noted a key distinction in terms of whom Palacios is being marketed to.
Port O’Connor is said to be attractive to, “people who do Gulf fishing, who want to be close to
the coast, who have boats and, and who like to fish.” Whereas Palacios is for, “people who like
to be on the water but don’t necessarily like to fish; people who are looking for a nice, laid-back,
relaxed place; who are less active.” Property tax rates were said to be “crazy” by a shrimp farm
manager who was interviewed, and he added that “the only people who can afford to buy on the
water are wealthy.”
118
Port Lavaca/Seadrift
Interviewees in Port Lavaca and Seadrift generally understated both the effects of
gentrification on commercial fisherman and the future use of tourism as an economic growth
strategy. Interviewees in Port Lavaca stressed the diversified nature of the city’s economy, and
in particular, its industrialization with the arrival of Alcoa and later Dow and Formosa as the
characterizing features. Virtually no interviewees described residential real estate development
or gentrification as a process having any considerable socio-economic or socio-cultural effects
on commercial fishers.
Several interviewees within and outside the town referred to Seadrift in terms of its
“backwardness,” as if it were a “poorer cousin” to Port Lavaca or “a small Midwestern city
whose time has come and gone.” Several interviewees from Seadrift seemed to fear similar
effects of real estate development such as those occurring in Port O’Connor and Palacios. One
bay shrimper noted, “Well, when these outsiders come in and buy up all the land, taxes is going
to go up and we will not be able to pay the higher taxes…we will no longer be a fishing
community.” Another bay shrimper stated, “Seadrift is changing. Shrimping, crabbing and
fishing is becoming a thing of the past. Sport fishing and expensive marinas seem to be the
future.” The president of the Chamber of Commerce, thought that Seadrift used to be a fishing
community, but that it is not anymore, and that the town is “going through a transition just like
many coastal towns I suppose.”
There are two housing developments taking place on the county side of Seadrift (part of
the ‘town’ of Seadrift lies within municipal boundaries and part within county boundaries) but
are using city utilities. These are primarily being bought by retirees and prospective secondhome owners. According to him, while the housing developments would likely increase the
value of property in parts of Seadrift, it would not increase to the extent that it would raise
everyone’s taxes. He noted that the Chamber supports the use of docks for sports fishing for an
increase in launching fee revenue, but this hasn’t taken off to an extent where former shrimpers
or fishers can hire themselves out as guides.
The geographic location of Port O’Connor and Palacios, as well as their respective socioeconomic histories, make them more amenable to the types of gentrification discussed by Smith
and Jepson (1993) and Blount (2006), amongst others. While not ignoring the particularities of
each locality, there appears to be considerable evidence of the ideal-type dynamics of
gentrification. These include (a) an increased prioritizing of recreational fishing and upscale
residential development for tourists and second-home owners that contributes significantly to the
inabilities of commercial fishers to obtain dock space; (b) the inability of commercial fishers to
obtain dock space and maintain homes due to the increase in property taxes; and c) a growing
cultural devaluation of their activities as wealthier persons move in and alter the social fabric of
one-time fishing communities. This dynamic seemed to be most pronounced in Port O’Connor.
119
Appendix D
Gentrification: Slated Development Projects
120
A. Calhoun County Economic Development Corporation
Current Developments: Any project that is complete or under construction.
1. The Sanctuary at Costa Grande (767 lots) is under construction. The plats have been filed.
Phase I has been sold and Phase II sales are underway. A WCID has been established.
Water will be provided through an interlocal agreement with the Port O’Connor MUD. This
project is located in Calhoun County.
2. Caracol (74 lots) is complete and sales continue. The project is located in Port O’Connor, an
unincorporated area of Calhoun County. Water is provided by the Port O’Connor MUD.
3. Swan Point Landing (89 lots) is complete and sales continue. This project is located outside
the City of Seadrift. Water is provided by the City via a single feed 6” line.
4. The Bay Club at Falcon Point (108 lots) is currently under construction. Three phases have
been recorded and sales are underway. Water will be provided by a public water system
utilizing shallow wells on the adjoining ranch property.
5. Seaport Lakes (56 lots) is complete and sales continue. Water is provided by individual
private wells.
6. Bay Pointe (99 lots) located in the county on Highway 316. Fifty-five lots are on the water
and some lots are 1, 5 and 10 acres. Roads are paved and electricity and GBRA water are in
place. A homeowners association has been established. There were contracts on all of the
lots within a 24 hour period in July 2006.
7. Blue Heron (38 lots) located on Highway 238. This project is an extra territorial jurisdiction
(ETJ) to the City of Port Lavaca. City water is available, but individual septic systems will
be required. All lots have been sold. A homeowners association has been established.
Planned Developments: A proposed project that has moved significantly toward construction
but for which neither a final plat has been filed nor have sales begun. Examples are submittals of
the Corps of Engineers permit applications, final planning and preliminary engineering.
8. Two Hotels and a Restaurant will be located on 14.04 acres on Texas Highway 35 and US
87. It is expected there will be space left over for a small strip of retail businesses.
9. The Sanctuary at Costa Grande Phase III (approximately 300 lots) is currently being
planned. Phase III will be located adjacent to Phases I and II.
10. Falcon Point Ranch (1,500 units) is continuing development and the Corps of Engineers
Permit application has been submitted. This project is included in an existing WCID. The
water source for this project will be either a pubic reverse osmosis treatment system, or an
inter-local agreement with the City of Seadrift, or a combination of both.
121
11. Harbor Mist (1700 lots) is located between the Victoria Barge Canal and Highway 185 in
Calhoun County. Canal permits have been issued by the Corps of Engineers. Phase I will
consist of 225 lots. Water will be provided by a proposed public reverse osmosis treatment
system operated by a WCID.
12. The Tidelands (approximately 82 units) is located in Port O’Connor (Calhoun County). The
Corps of Engineers permit approval is expected soon and platting and engineering design
have begun. Water will be provided by the Port O’Connor MUD.
13. Powderhorn Ranch is currently planning and has begun preliminary construction of a golf
course. The current plan calls for a low density residential project containing up to 500
units. The ranch is located adjacent to Port O’Connor, Calhoun County. Water would be
provided by individual private wells.
Potential Developments: Property owned by individuals or entities that are actively considering
developing the property.
14. The Sanctuary at Costa Grande consists of about 10,000 additional acres that are slated for
the development of possibly 8,900 units.
15. Falcon Point Ranch owns 4,000 acres of additional developable property but no planning is
currently underway.
16. Powderhorn Ranch has significant additional property with extensive water frontage. No
plans are currently being pursued.
17. The Bindewald Tract (300 lots) is currently being planned as a low density project and
would probably rely on private individual wells as a source of water. This tract is located
between Highway 185 and the Victoria Barge Canal.
18. The Fisher Tract (300 lots) is currently being considered for a canal lot sub- division. This
tract is located just outside of the City of Seadrift. The 6” waterline that feeds Swan Point
Landing passes through this property. The City of Seadrift has committed to a small
number of taps as a condition of easement dedication; however, this is not nearly adequate
for the number of units planned. Additional water sources would include an amended
agreement with the City of Seadrift or a reverse osmosis system.
19. Lane Road is an undeveloped area that is being divided into 5 acre tracts. There is a
possibility of up to 300 residential units in this area that will be served by individual private
water wells. This tract is next to the extensive acreage of Costa Grande on Highway 185
outside of the City of Seadrift.
122
Announced Projects :
Formosa Plastics Corp TX - $90 million specialty PVC plant
Permanent jobs = 100; Construction jobs expected = 500-600 for 2-3 years
Formosa Plastics Corp TX – Pet-coke plant.
Permanent jobs = 60; Construction jobs expected = 250-350 for 3 years
Calhoun LNG – two 160,000 cubic meter storage tanks and 1 Bcf/day gas vaporization and
natural gas liquid separation capacity terminal.
Permanent jobs = 50; Construction jobs expected = 400-600 for 3 years
NuCoastal Power – Pet-coke fueled plant.
Permanent jobs = 80; Construction jobs expected = 300 for 3 years
INEOS Nitriles - $90 million for installation of 4th reactor train to increase plant capacity 15%
Construction jobs expected = 300-600 until fall 2008
Seadrift Coke LLP – 40% plant expansion in 4 phases with phase 1 completed
Permanent jobs = 10; Construction jobs expected = 200-300
Excalibur Processing, LLC - $140 million hydrocarbon processing facility. Construction
expected to begin late 2008.
Permanent jobs = 42; Construction jobs expected = 200-300 for 2 years
LaQuinta Motel - $3 million project. Construction began July 2007.
Permanent jobs = 7-8
Kasan-Ringgold Investments LTD / International Flight – To build 2 motels, a restaurant and a
small retail center
Projects in Negotiation Process:
Chemical Facility - $450 million
Permanent jobs = 650
Assisting local industry to bring a major customer to Calhoun County to build a new plant
Container facility on Victoria Barge Canal connecting to Intracoastal Waterway
Proposal Submitted
Project DeMarco – Manufacturer of natural, organic food products. $4-5 million in machinery
and equipment;
Permanent jobs = 25; Site visits July-August 2008.
123
Education/Training
Victoria Junior College opened a full course curriculum center in Port Lavaca in June 2007. One
major asset of this center will be workforce training opportunities and specifically designed
courses for industry.
B. Galveston Bay Fishing Communities
Place
Project
Location
On Eckert
Bayou, midisland
500 Seawall
Blvd
Galveston
Laffite's at Pirate's
Cove
Galveston
Emerald by the Sea
Kemah
15-story tower on east end;
$375k-1.5m
Park at Waterford
Harbor
1420 Marina
Bay Dr.
Gate rental apartments; $780$1,583/mo.
Kemah
Grand Bay at
Kemah
Waterfront
across 146
90-acre luxury single family
and multi-family, upscale
retail, banks, boutique hotels,
and marina on Galveston Bay
San Leon
Gordy Park
Waterfront
across 146
Seabrook
Wreckers
(restaurant)
FM 517 & 20th
St.
Texas City
Lago Mar
N. of FM 1764
to Holland Rd
Texas City
Sandpiper
146 & 25th
4,000 residential units near
Bayou Golf Course
Texas City
Grand Cay Harbor
End of Skyline
Drive @ GB
580 home lots
Texas City
Lone Trail Village
I-45 & Century
Blvd
225-home development;
Gehan Homes
W of FM 3436,
b/w 517 & 646
1,200 home community;
Dallas developer John Marlin
Texas City
124
Notes
35-unit condo near Galveston
Country Club; $760k/3br unit
8,000 sq. ft, "tropical flair",
w/dockside boat parking &
ramp
Homes from $140k-$1m; will
include office space/retail;
massive property tax re
investment will subsidize
supporting infrastructure
including a lake
Appendix E
Impacts of Hurricane Ike
125
Hurricane Ike made landfall on September 13, 2008. Although it was classified as a
Category 3 storm due to its projected winds, its storm surge was projected to be a Category 5.
This discrepancy confused many people, and combined with memories of a disastrous
evacuation of Hurricane Rita, many residents chose to stay.
Figure 7: Path of Hurricane Ike.
Damage to Fisheries
The exact damage to fisheries is not known at this time. Texas Sea Grant has distributed
a survey to the fishing industry in an attempt to get detailed information about losses. This
information will be available to industry to aid it in securing funding for rehabilitation of
infrastructure, boats and ecosystems. The information below is from post-Ike conversations with
fishermen, TPWD, Sea Grant and news reports.
Oysters
Hurricane Ike caused significant damage to the oyster reefs both on the eastern and
western shores of Galveston Bay. Initial side-scan sonar data collected following the storm was
compared with data collected before the storm and indicated that approximately 50-60% of the
oyster reefs in Galveston Bay were lost due to siltation/sedimentation. East Bay (a sub bay of
the Galveston Bay system) was the hardest hit, with losses in excess of 80%. The public reef
fishery that normally opens November 1 was delayed for three weeks due to the loss of markers
designating the boundary between approved and restricted waters. It is unlikely that the newly
damaged reefs will return without intervention. TPWD will apply for funds for reef restoration,
but this will take years. Production for some harvesters/leases is down 80% according to the
owner of one of the three major oyster companies on the bay.
126
Figure 8: Damage in San Leon from Hurricane Ike. Photo: Lisa Gonzalez
A significant proportion of oyster infrastructure is located in San Leon, a community
which was hit very hard. Dock owners face damaged docks and have ownerless boats on their
docks. Texas law dictates that owners be notified before removal, further complicating dock
repair. Several of the large oyster leaseholders also have leases and facilities in Louisiana. They
were able to save most of their boats by taking them to their other facilities and have been
harvesting in Louisiana since the storm.
The day before the public season was set to begin, one oysterman stated that his biggest
problem since the storm has been finding available crew. His son, currently a college student in
Texas City, will help him temporarily until he can find reliable deckhands. He sustained $3000
in damage to his boat as a result of the storm. He said it would have been much more except that
a friend was able to do the necessary welding work.
Crabbing
The Galveston Bay crab fishery is dominated by Vietnamese fishermen and is
concentrated in San Leon (one of our research communities) and Oak Island (not one of our
research communities). Both communities suffered some of the most severe damage from the
storm. Oak Island was almost completely destroyed and an estimated 1,500 of 1,815 homes in
San Leon sustained severe damage. Vietnamese interviewees estimated that 85% of Vietnamese
in San Leon have lost their homes which were largely paid for in cash with little flood insurance.
The crab processing plant in San Leon was still closed at the time of writing this report
and the owner indicated that she did not know when it might re-open. Boats and crab traps were
destroyed.
127
Shrimping
Figure 9: Damage in San Leon due to Hurricane Ike. Photo: Lisa Gonzalez
The Mosquito Fleet on Galveston Island lost four boats due to Hurricane Ike. Two were
sunk, while two others were washed aground by the surge. The interviewer was told that these
four boats were owned by fishermen who did not go out regularly and they did not take
necessary precautions prior to the storm. This appears to be the case in other parts of the Bay.
Boats in San Leon that were unused last season were damaged while boats being used were taken
to safety. Boats from Bolivar, Texas City, and a few Galveston boats that had previously docked
near the Causeway between Galveston and the mainland are now docking at Pier 19.
There are two Vietnamese fish houses in San Leon that handle shrimp. Some fishermen
were able to get their boats to safety, but many were not. Shrimp season was delayed due to
water quality concerns and debris. Despite these problems, shrimp season did re-open.
Bait camps all around the bay were damaged and not all were insured. Those along the
Texas City dike, where several of our interviewees sold bait, were destroyed and the dike was
“wiped clean.” At this point, the city does not plan to allow these camps to rebuild on the dike.
Some might be able to re-build on the shore but at this point, it is unknown.
In October of 2009, FEMA allocated 5 million towards dike repair, falling short by an
estimated 6.3 million of what is needed (Aulds, 2009). Repairs will focus on land based
amenities and will not include fishing structures such as bait camps and piers. It is estimated that
fishing infrastructure will not be repaired for another year.
128
Figure 10: Damage to Texas City Dike from Hurricane Ike. Photo: Lisa Gonzalez
One employment opportunity that will be available to fishermen is with TPWD. The
agency has secured funds to hire fishermen to help with debris removal.
Recreational fishing
According to TPWD assessments, 60 of the 127 boat ramps on Galveston Bay that
provide access to recreational fishermen (and other boaters) were damaged. This represents a
significant cost to the state because the average cost of repairs to boat ramps after Hurricane Rita
was $125,000 per ramp.
Hurricane Ike Recovery at the Community Level
Galveston
The post-Ike recovery efforts in Galveston have progressed in recent months from
meeting immediate needs such as the restoration of basic services and debris removal to more
long-term concerns regarding local economies and housing refurbishment. Federal recovery
funds are now in the pipeline for the Galveston Bay area. In February, the Houston-Galveston
Area Council announced its ratio for the distribution of $814m in Community Development
Block Grant (CDBG) funding. All jurisdictions who receive the funds are required to
demonstrate concrete action plans for the implementation of the funding within two years. There
are several recovery institutions committed to serving both the city and county of Galveston.
These include the Galveston Long Term Recovery Committee, the Galveston County Restore
and Rebuild Coalition, and the Northside Galveston Taskforce.
129
The goal of the Long Term Recovery Committee is to formulate tangible recovery
projects with the technical assistance of individuals who are contracted with FEMA. The
Northside Galveston Taskforce is a volunteer organization affiliated with GRACE Friends
Services Inc., the first African-American based case management organization in the county.
The organization is focusing its efforts on the restoration of public housing, the return of
displaced residents, the assurance of health care either through UTMB or St. Vincent’s Health
Clinic, community enrichment projects such as after-school and cultural centers, and
partnerships with local schools and churches.
The Galveston County Restore and Rebuild Coalition (GCRR) emerged in October
through the organizing efforts of an ex-Galveston mayor and community organizers. GCRR is
comprised of approximately twenty faith-based and/or non-profit entities that focus on housing
rehabilitation and meeting unmet needs.
Housing
Hurricane Ike flooded 75% of Galveston Island with a storm surge estimated at between
17 and 20 feet.
The vast majority of homesteads assessed by the City of Galveston and FEMA as
substantially damaged or destroyed as of December 2008 are located behind the Seawall. These
were homes that represented the affordable housing sector on the island.
Average Homestead Value for entire city:
$ 135,084.64
Average Homestead Value for Substantially Damaged:
$ 62,918.70
Post-Storm Transitional Housing
FEMA has set up a mobile home park consisting of 54 units near Scholes Airport. It is
currently not known how many individuals and families will ultimately be awarded temporary
mobile homes. There are currently 180 families on a waiting list. Unless the City extends the
deadline, FEMA would be required to remove the units by April 30, 2010. Five months after
Ike, there were still approximately 1,200 Galveston County families living in hotels. Unless
FEMA once again extends their occupancy, they will be forced out on March 13. At the time of
the writing of this report, it is unclear whether displaced Galvestonians currently residing in
hotels/motels will be awarded one of the 36 available mobile home units at the group site. In
addition to the trailer park, FEMA estimates that there are currently 162 mobile units on private
properties within the City of Galveston, and 346 scattered throughout Galveston County.
Currently, 293 private landlords on the island have expressed their willingness to accept
vouchers.
Disaster Housing Assistance Program (DHAP): HUD provides the Galveston Housing
Authority the names of County residents who qualify for rental assistance until repairs to their
130
homes are completed. The GHA then provides vouchers to residents to find rental apartments or
houses through landlords who are willing to accept them. This new tenant is then responsible for
forwarding the rental agreement back to HUD, who will in turn pay the landlord the monthly
rent. However, according to the Daily News, a perception is circulating amongst area landlords
that the GHA has been missing payments for DHAP and displaced Section 8 families. Other
landlords have been turning applicants away on the grounds of poor credit and concerns they
won’t be able to pay fair market rent once government assistance expires. This dynamic has
been particularly acute for displaced residents of local public housing. Consequently, FEMA has
been providing “bridge payments” to residents until the GHA resumes rental payments.
Furthermore, as noted above, certain recipients of DHAP have been unable to find landlords who
will accept their vouchers on the grounds that they cannot provide sufficient income or credit.
Public Housing
There were approximately 580 families in public housing that were displaced by
Hurricane Ike. In early February, the Galveston Housing Authority announced that it would
demolish two of the four public housing developments on the island. Oleander Homes and Palm
Terrace will be razed, while Magnolia Homes and Cedar Terrace will be rehabilitated as
transitional housing units while new projects are developed over the next several years with
Community Development Block Grant funds. The Housing Authority is now soliciting bids to
complete the renovations and it is now scheduled to receive displaced residents by the fall.
According to the Executive Director of the GHA, it will take at least two years to replace the
other two housing projects that were demolished. Federal funding for the reconstruction of the
public housing units must be mobilized two years from the disaster event but there are no firm
development plans for the new projects. Members of the Northside Galveston Task Force have
expressed significant concern over the future of public housing, and some are concerned that the
housing units will not be replaced, leading to the permanent displacement of the island’s poorest
and most vulnerable residents.
Housing Rehabilitation
The housing subgroup of the Long Term Planning Committee is currently developing
three projects. One of these will be a comprehensive housing needs assessment that shall
identify current needs, available housing stock (single and multi-unit), vacant lots,
abandoned/delinquent properties, and units that are in code violation.
The needs assessment study will then inform the targeted implementation of two housing
programs: Housing Rehabilitation and Existing Neighborhood Infill; and the Sally Abston Rentto-own/Work-to-own Homes Program. The first program has established their goal to “save
existing neighborhoods and create affordable, clean, safe, attractive housing at a variety of pricepoints based on the results of the needs assessment.” The program would utilize CDBG and
other HUD program funds in conjunction with labor-power from non-profit organizations to
131
provide rehabilitation assistance to homeowners whose houses were substantially damaged by
Ike. The infill segment will allow non-profit and for-profit developers the ability to purchase
city-owned vacant lots and construct new-homes for low-to-moderate income families at a sales
price that would not exceed $99k. The housing design would result from a design competition.
In Galveston, the goal would be to award a competitive design and build homes that could be
bought at or under $125,000.
The goal of the Sally Abston program is to “attract hard-working low/middle income
residents of Galveston to become homeowners through a self-help construction program that
provides options for financing, including lease-to-own.” It will work in tandem with inter-faith
groups and a newly created Galveston chapter of Habitat for Humanity. Ideally, the program
would build homes with a sales price of $40-80,000. Financing would be facilitated through a
basic market loan leveraged with low-or-no interest loans that come through CDBG funds. A
Housing Trust Fund created, theoretically, by the City, would also provide down payment
assistance for select residents. For both home-building programs, FEMA recommends outreach
to at least two National Certified Housing Development Organizations as primary developers.
The Galveston County Restore and Rebuild Coalition is also working on housing
rehabilitation. It will allocate up to $15,000 per household for home repair efforts that include
cleanouts, assessments and estimates, case management of donated materials, assignment of
work crews and construction management. Clients are expected to use personal resources such
as FEMA payment and any insurance payments to aid in their rebuilding efforts. Unmet Needs
Funds of up to $1,500 per qualified client or family are distributed for items and services deemed
necessary for recovery of personal property. Case managers present requests to the Unmet
Needs Committee on a weekly basis. The committee also receives in-kind donations and works
with case managers to distribute appropriate items to clients.
Prior to the storm, the major economic engines of the City of Galveston were tourism and
the University of Texas Medical Branch (UTMB).
The University of Texas Medical Branch (UTMB)
UTMB was a major economic force in Galveston and the announcement that a significant
portion of its functions will not be restored was considered by many to be the final blow
delivered by Ike. As of this writing, the future of UTMB remains in doubt. The Board of
Regents proposed to take a significant amount of medical care and jobs to the mainland (this is
described below) but has modified their earlier position due to a prolonged and concerted effort
by islanders to keep the facility. Currently, an agreement has been struck that the facility will be
restored to pre-Ike functions if the Texas Legislature will supply a significant financial match.
Thus, the fate of the clinical services is still in question.
132
UTMB laid-off about 3,000 workers and curtailed patient services. Although UTMB did
not operate the Shriners Hospital for Children, the two institutions were closely intertwined. In
late January, Shriners Incorporated announced that it would suspend storm-restoration on the
hospital as a cost-cutting strategy in the face of a $3b endowment shortfall brought upon by the
collapse of financial markets. This expedited another 325 lay-offs.
On February 11, two competing documents were released to the public hours apart, both
of which demonstrate the divergent pathways that the institution will take over the next several
years. At the behest of Gov. Perry, the UT Board of Regents commissioned a consulting firm
specializing in healthcare facility planning to conduct a study that assessed the long-term
economic viability of the UTMB according to scenarios ranging from pre-storm restoration to
full-transference inland. The firm recommended that UTMB transfer its hospital facilities to
League City, while retaining only its state prisoner-patient population and the medical school
facilities on island. Prior to the storm, UTMB began building a $61 million specialty care center
in League City. Also just before the hurricane, the University paid $9.4 million for 29 acres near
the specialty center, a likely spot for a hospital.
The report projects significant operating losses even if the medical branch were to move
hospital operations to League City, but that losses would be mitigated in the long-term as UTMB
establishes a consistent revenue base in the north county amongst a greater proportion of insured
and Medicaid-backed patients.
An alternative vision, that is backed by Galveston residents but does not seem to be
favored by UT Regents would leverage federal, state, and charitable funding to move ahead with
plans to build a $250 million surgical tower on-island to bring a total of 528 hospital beds to the
island, harden John Sealy Hospital to prevent future hurricane damage, and restore the medical
branch’s Level 1 trauma center. Restoring the hospital is generally viewed as necessary for both
the economic and physical health of island residents.
Economic Redevelopment
The Long Term Recovery Committee instituted a working group focused on economic
development. The group has largely harnessed development plans that pre-existed the storm,
including the Downtown Redevelopment Plan that had been sponsored by the Historical
Downtown Galveston Partnership (HDGP); the development of the East-side Apfell Park as an
eco-tourist attraction; a business incubator; Port expansion that includes expanding the berths for
cruise ships and the expansion of facilities on Pelican Island; and also the realization of the
Seawall Master Plan that blends concepts and projects from the 1983 and 2006 plans.
The HDGP has secured slightly more than half of the approximately $400,000 necessary
to commission to Downtown Redevelopment Plan. The Plan seeks to develop the harbor side
from approximately Pier 26 eastward to Pier 10 as a mix-use development that would include
retail and residential. This may be facilitated by shifting industrial operations currently in that
133
zone further north or onto Pelican Island, said one member of the working group and a leader of
Historical Downtown Galveston Partnership. Currently, there are no plants to displace the
commercial fishing Fleet that uses piers in the redevelopment zone.
According to the Park Board, tourism in 2007 had a total economic impact of over $800
million and supported 11,500 jobs on the island. The Galveston beaches received about 6.5
million visitors each year, and contributed more than $705 million to Galveston Island in 2006
alone (AngelouEconomics, 2008). The beaches were badly damaged by the storm and beach
replenishment represents a major funding request. There is currently a major beach
reconstruction project occurring from between 61st to Stewart Beach and Park Board members
are searching for a dedicated source of revenue to support the beaches’ maintenance and
amenities.
Another significant generator of tourism dollars are several festivals occurring throughout
the year. The earliest of these, “Dickens on the Strand,” occurred just 3 months after the storm
despite the significant damage that the Strand suffered. Mardi Gras was held a few months later
and plans are now being made for the Lone Star Motorcycle Rally.
Kemah
Estimated losses in Kemah include at least:
- 202 residential units substantially damaged or destroyed
- $100 million in commercial damage
- $2.4 million in damage to infrastructure and facilities
- 1,700 employees out of work
- $100,000 in sales tax lost per month
According to Kemah’s City Manager virtually all debris has been removed from within
the city limits and recovery efforts are now devoted towards street repaving (an issue pre-Ike)
and the replacement of the local MUD’s pump station. 97% of the city’s operational revenue
comes from sales tax, and the main economic engines are the Boardwalk, WalMart and Target.
The Boardwalk is operational and the marina next to the Boardwalk was constructed to float
during the storm and a visual inspection by the research team post-storm indicated that only one
boat was lost. However, other businesses are not yet functioning and the city expects a budget
shortfall this year.
The most substantial damage to the housing stock occurred on the west end of the city,
where the population is largely Hispanic. According to the city manager, this land is too
valuable for Hazard Mitigation buyouts and may be an area that will gentrify over the next
several years. The majority of displaced residents in Kemah come from the west-side.
134
Seabrook
Estimated losses in Seabrook include at least:
- 1,966 residential units that were substantially damaged or destroyed
- $35m in commercial damage
- $5.7m in damage to infrastructure and facilities
- $127k in lost tax revenue
Seabrook city officials, including members of the Economic Development Corporation
(EDC), met to discuss possible redevelopment options for Old Seabrook and The Point (where
the Seabrook seafood wholesalers are currently located on Waterfront Drive). Of the eight
seafood wholesalers on Waterfront Drive, four are now open: Golden Seafood, Emery’s, Pier 8,
and Waterfront Seafood.
The Seabrook EDC has applied for grant funds to develop Waterfront Drive.
meeting was held for discussion purposes only and no action was taken.
The
The Waterfront Development Plan included the following priorities for the development of
the Point:








Develop a central green space around which commercial activity can thrive.
Provide uninterrupted public access to the waterfront perimeter of the Point.
Create a habitat and recreation island that offers the Point and lower Todville Road
protection from storm events.
Connect the Point to the Marina District and Old Seabrook via pedestrian and multi-use
trails.
Expand Seabrook’s marina facilities including both resident and transient dockage by
creating a dredged basin that can accompany between 300-350 slips.
Enhance walkability.
Develop the Point as a mixed-use development.
Maintain the Fishing Fleet as a vital asset to the community.
The Plan excludes housing development on the Point, instead suggesting that the area can
accommodate restaurants and fish markets to attract tourism, within the context of connection to
the marina district. This includes plans for the creation of a habitat island and a raised boardwalk
that connects it to the Todville Road/Waterfront Drive district. Additionally, multi-use trails will
extend parallel with Todville. According to the document, “the recommendations in this plan
aim to create a unique environment that combines the working waterfront of the shrimping and
seafood industry with a blend of retail, commercial, marina and recreation activities” (23).
Further, the plan also stipulates that:
135
“Every effort should be made during redevelopment efforts to retain the working shrimp fleet
and associated retail seafood markets. These markets offer a uniqueness and authenticity to the
Point that cannot be found in similar nearby waterfront districts. Through thoughtful site design
and an understanding of the needs of the seafood markets, retail operations and customers, a
clean and safe environment can be created. Examples of this mix of working waterfront and
public retail development include GranVille Island in Vancouver, Seattle’s Pike Place Market
and San Francisco’s Fisherman’s Wharf” (Ibid).
San Leon
The unincorporated village suffered a storm surge of 10-13 feet. San Leon was one of
the worst hit of the Galveston communities. An estimated 1500 out of 1800 houses are
uninhabitable with about 850 completely destroyed. About 65% of San Leon residents stayed
during the storm, resulting in several deaths. Despite the heavy losses, the honorary “mayor” of
San Leon (elected during the annual “Where the Hell is San Leon” festival) sent a plea to
residents to return to start self-help restoration shortly after the storm. San Leon offers a good
example of the power of social capital and self reliance. A local restaurant hired back its waiters
for clean up and then distributed free meals. Churches responded rapidly and served as
distribution points before FEMA arrived. Early outside aid included the Red Cross and the Lions
Club. There are some FEMA trailers scattered throughout the neighborhoods.
Figure 11: Damage to a Residential Area in San Leon from Hurricane Ike. Photo: Lisa
Gonzalez
136
The local Community Church still holds food distributions every Friday morning in
conjunction with the Houston Food Basket. Volunteers have noted a steady decrease in volume
over the last two months. Based on a visual survey of the area in late February, much of the
visible debris has been removed and there are noticeably more empty lots as a result of
demolition. Because of its lack of incorporation, formal planning such as that seen in other
towns is not taking place.
TWIA
The Texas Windstorm Insurance Association has received 90,656 reported losses and
paid $905 million in claims. But 10,500 claims are unresolved either because they involve very
large and complicated commercial cases or are mired in the murky question of whether wind or
storm surge caused the damage. Approximately 3,000 windstorm claims were for total losses;
where nothing was left of the insured structure but slabs or sticks. The claim resolution process
has been particularly contentious since the majority of Ike’s destruction resulted not from the
110mph winds, but from the 17-20 ft. storm surge; thus, bogging down the resolution process
with the murky question of whether damage was done by wind or flooding. The problem has
reportedly been particularly acute on Bolivar. With high proportion of total-loss claims on the
Peninsula, adjusters have been forced to rely heavily on computer modeling that is based on
structural information collected from houses still-standing to determine the wind damage.
As of Feb. 9, there had been 1,007 complaints filed against TWIA. Of those, the
insurance department deemed 382 “justified,” resulting in the return of $5.9 million to
policyholders. The top reason for complaint to TWIA was “slow-settlement.”
137
Appendix F
Recreational and Commercial Fishing Infrastructure
138
Figure 12: Changes in Access sites since 1990 – Southern Research Sites.
139
Figure 13: Changes in Access sites since 1990 – Northern Region (does not include hurricane
damage).
140
Figure 14: Locations of Kemah and Seabrook Docks and Fish Houses.
141