DEEPWATER HORIZON OIL LEAK: A DECISION ANALYTIC APPROACH TO RESOURCE ALLOCATION A DISSERTATION SUBMITTED ON THE TWENTY NINTH DAY OF FEBRUARY 2012 TO THE DEPARTMENT OF ENVIRONMENTAL HEALTH SCIENCES IN PARTIAL FULFILLMENT OF THE REQUIREMENTS OF THE SCHOOL OF PUBLIC HEALTH AND TROPICAL MEDICINE OF TULANE UNIVERSITY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY BY BENJAMIN CHARLES SCHULTE DOCTORAL COMMITTEE APPROVED: PETER FOS, P . ., DDS, M.P.H. SARAH K. MACK, Ph.D., M.S.P.H. COMMITTEE CO-CHAIR /4YJI;L " ROBERT REIMERS, Ph.D.
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
DEEPWATER HORIZON OIL LEAK: A DECISION ANALYTIC APPROACH TO RESOURCE ALLOCATION
A DISSERTATION SUBMITTED ON THE TWENTY NINTH DAY OF FEBRUARY 2012
TO THE DEPARTMENT OF ENVIRONMENTAL HEALTH SCIENCES IN PARTIAL FULFILLMENT OF THE REQUIREMENTS
OF THE SCHOOL OF PUBLIC HEALTH AND TROPICAL MEDICINE OF TULANE UNIVERSITY
FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY BY
·~~ BENJAMIN CHARLES SCHULTE
DOCTORAL COMMITTEE
APPROVED:
PETER FOS, P . ., DDS, M.P.H. SARAH K. MACK, Ph.D., M.S.P.H. COMMITTEE CO-CHAIR
I would first like to extend my gratitude and appreciation to my dissertation committee. My advisor and committee co-chair, Dr. Andrew Englande, Jr., for his exceptional guidance and dedication in professional and personal mentorship over the duration of the program. Dr. Peter Fos, for his advice, support and encouragement throughout the dissertation process. Dr. Sarah Mack, for her practical and insightful perspective in approaching the dissertation. Dr. Robert Reimers for his consistent availability and motivational talks. Lastly, I would like to thank Dr. Janet Rice for her availability and assistance with the statistical portion of my dissertation.
I would like to extend my greatest gratitude towards my family and friends: My mother Mary Bechtel Schulte for her countless hours of editing, support, and love during this dissertation and throughout my life. My father, Fred Schulte and step-mother Sue Chattier Schulte for their continued support and encouragement. My brother F. Martin Schulte and sister-in-law, Thea McMahen for spurring me on in moments of doubt. To all my friends with specific thanks to James Duddy, Ester Kim, Piero Spadaro, Mikiala White, Dr. Thomas Vander Jagt, Ashley McConnell Vander Jagt, Diana Hamer, Drs. J.B. and Deborah Barbeau. A very special thanks to my Masters advisor, Dr. William Toscano, Jr. for his continued support extending well beyond my time at the University of Minnesota.
2
ABSTRACT
On AprillO, 2010, The British Petroleum (BP) Deepwater Horizon oil-drilling rig
experienced a blowout resulting in the death of eleven rig workers and an oil leak of 4.9
million barrels into the Gulf of Mexico that lasted eighty-six days. It was the largest oil
spill related incident to occur in United States coastal waters. The immediate impacts of
the leak were extensive to the ecosystem, wildlife, as well as the communities dependent
on the environment and oil and gas operations for their livelihood. Moratoriums were
placed on deepwater drilling and commercial fishing operations as precautionary
measures to protect human health and to re-evaluate drilling operational procedures.
Although the full impacts associated with the Deepwater Horizon incident are still
being assessed and litigated, the incident caused harm to the coastal region communities.
BP 's lack of preparedness to handle an incident like the Deepwater Horizon event
provides strength to the idea that impacted communities must be self-reliant in preparing
for future disasters and that funds resulting from federal penalties levied against BP
should be allocated towards the communities adversely impacted by their negligence.
This idea was given additional credence by the National Oil Spill Commission's
recommendation that 80% of the Federal Clean Water Act fines be allocated towards
environmental, economic, and societal resilience measures in impacted regions in their
fmal report. The report further identifies a need to prioritize tasks and make the
recommendations binding.
This research identifies and prioritizes resilience measures in St. Bernard Parish, LA
with consideration to future oil related measures. This study developed a decision
analytic tool that creates a framework for prioritization of monetary resource allocation
that promotes long-term resilience of St. Bernard Parish with consideration to future oil
related events. The model incorporates impacts associated with the oil leak in the
recovery stage as well addresses sustained resilience deficiencies in the Parish. The study
defined long-term as the time at which initial distribution of civil penalties associated
with the Deepwater Horizon event are allocated to flfty years in the future. The Delphi
3
method identified thirty-three model variables through expert input and consensus. The
created multi-criteria decision model is based on the Simple Multi-Attribute Rating
Technique (SMART) methodology. SMART determined expert value trade-offs of model
variables through a scoring system based on a ratio assessment procedure and they were
listed from highest trade-off scores to the lowest.
The results indicated a high prioritization for monetary resource allocation of
variables associated with environmental and economic stability. Variables associated
with increasing the capacity to respond to future oil related disasters ranked particularly
low in the prioritization of monetary resource allocation. The identification and
prioritization of model variables indicate a strong sentiment among expert participants to
implement more generalized resilience measures that will prepare the Parish for future
man-made and natural disasters, opposed to specifically bolstering resilience measures
pertaining to future oil related events in the Gulf of Mexico.
The thirty expert participants were placed in four expert groups: Science and
Technical representatives, Business representatives, Government representatives, and
Community Based organization representatives. To assess the experts value trade-offs of
the model variables, Kendall's Coefficient of Concordance was used to identify the level
of agreement among experts as a whole and within their respective expert group.
Agreement in ranking model variables was found for twenty-nine of thirty-three variables
assessed. The levels of agreement were higher in the individual groups when compared
to the expert group as whole. The low to moderate levels of agreement within the
individual groups indicate diverse perspectives and value trade-offs by experts in
prioritizing the identified resilience measures. The Kruskal-Wallis H-test was used to
assess agreement in median ranking of model variables between expert groups.
Agreement in median rankings was identified in twenty-five of thirty-three model
variables among the four expert groups. Disagreement was found in ranking of five
environmental variables.
This research recommends the use of civil penalties associated with the BP
Deepwater Horizon incident to implement the resilience measures identified and
prioritized in this research. A stipulation prior to implementation, however, is to conduct
4
a thorough cost assessment of each resilience measure and account for projects in St.
Bernard Parish with secured funding that are found in the prioritized list within this
research. The results of both assessments will provide guidance to policy makers to
TABLE OF CONTENTS .................................................................................................. 6
CHAPTER 1: Introduction .............................................................................................. 8 1.1 Significance and Background of Study ..................................................................... 8 1.2 Statement ofthe Problem ........................................................................................... 9 1.3 Purpose ofthe Study .................................................................................................. 9 1.4 Research Questions .................................................................................................. 11 1.5 Assumptions of the Study ........................................................................................ 11 1.6 Definition ofTerms ................................................................................................. 12
2.5 Application of Multiattribute Decision Modeling to Oil Related Incidents ........... .43 2.5.1 Analytic Network Process and Resource Allocation ...................................... .44 2.5.2 SMART and Resource Allocation .................................................................. .45
CHAPTER 3: Research Methodology .......................................................................... 47 3.1 Model Development ................................................................................................ 47 3.2 Structuring Decision Model ..................................................................................... 51 3.3 Survey Development. ............................................................................................... 52
4.4.1 Kendall's Coefficient of Concordance (W) Results ........................................ 73 4.4.2 Kruskal-Wallis H-test Results ......................................................................... 79 4.4.3 Hypothesis Test Conclusion and Summary ..................................................... 82
4.5 Prioritized List of Resource Allocation and Cumulative Percent List .................... 84 4.5.1 All Expert Response ........................................................................................ 85 4.5.2 Group Responses ............................................................................................. 88
4.6 Conclusion of Research Results ............................................................................. 98
CHAPTER 5: Discussion .............................................................................................. 102 5.1 Overview ofResults ............................................................................................... 102 5.2 Hypothesis Testing ................................................................................................ 106 5.3 Review of Objective Ranking ................................................................................ 107 5.4 Review of Environment Objective and Sub-objectives and Results ..................... 109 5.5 Review of Logistical Capacity Objective and Sub-Objectives and Results .......... 110 5.6 Review of Economic Objective and Sub-Objectives and Results ......................... 110 5.7 Review of Societal Impacts Objective and Sub-Objectives and Results ............... 111 5.8 Policy Implications ................................................................................................ 112 5.9 Public Health Pertinence ..............................•......................................................... 116 5.10 Conclusion and Summary .................................................................................... 118
The global demand and use of oil as an energy source has led to extensive searches
for oil reserves. Offshore exploration and extraction of oil is common worldwide and the
United States Department ofEnergy and Department oflnterior anticipate significant
growth ofthis practice in the future (Department ofEnergy, 2011). The United States
Outer Continental Shelf operations in the Gulf of Mexico yield 379,820 thousand barrels
per year and account for thirty percent of oil production in the United States (BOEMRE,
2011). Though offshore drilling is an economically viable process to extract oil, there are
associated ecological and subsequent societal risks should an incident occur.
On April10, 2010, the semi-submersible Deepwater Horizon oil drilling rig operated
by British Petroleum (BP) experienced a blowout which subsequently cost the lives of
eleven employees and sank the drilling rig. The leak continued for eighty-six days
releasing 4.9 million barrels of oil into the Gulf of Mexico. The short-term impacts of the
incident were significant in Gulf coast communities. Concerns associated with the safety
of oil drilling rig operations as well as the lack of preparedness and capacity by British
Petroleum to effectively respond the Deepwater Horizon incident resulted in a Federal
mandated moratorium on deepwater oil drilling in the Gulf of Mexico. Concerns related
to the contamination and bioaccumulation of oil in aquatic gulf species commonly
consumed by humans as well as the potential implications that large quantities of oil
would have on the complex and fragile Gulf ecosystem resulted in a moratorium on
fisheries and deepwater oil drilling operations. Due to the importance of oil and gas and
8
fisheries in coastal communities livelihood, the economic impact of the moratoriums
were significant to the well being of coastal populations (Austin, et. al., 2008)
As any disaster or major event, the Deepwater Horizon incident was unexpected. As
evidence of the duration and quantity of oil leaked, the responsible party's (British
Petroleum) preparedness plan to handle such an incident was insufficient. BP's
negligence highlights the need for coastal Parishes in Louisiana to be prepared and self
reliant to respond to any future event of this nature. Ensuring adequate preparation and
capacity to respond to such an event will greatly mitigate the impacts on a fragile
ecosystem and the economic stability of coastal communities.
Resilience is a principle concept in this research, thus it is important to defme the
concept. Though organizations define resilience differently, the fundamental meanings
are the same. The Community and Regional Resilience Institute defines resilience as
"The capability to anticipate risk, limit impact, and bounce back rapidly through survival,
adaptability, evolution, and growth in the face of turbulent change".
In addition to the importance of Parish preparedness to mitigate the short term
impacts of future disaster or incident, the Deepwater Horizon incident further highlighted
the lack of resilience in coastal Louisiana and its susceptibility to disasters. The
extensive canal and pipeline systems utilized by the oil and gas industry have resulted in
a high rate of coastal erosion and environmental degradation of wetlands. This is critical
as the coastal ecosystem and wetlands provide an important natural buffer against
hurricanes.
9
The historical connection between oil and gas operations and environmental
degradation in coastal Louisiana as well as British Petroleum's negligent operations and
preparedness plan to handle a drilling incident, provides an argument that a portion of the
penalties associated with the Deepwater Horizon incident be allocated toward the
promotion of resilience in areas impacted by a large scale oil leak. This concept of
resource allocation has been backed by the National Commission on British Petroleum
Deepwater Horizon Oil Spill and Offshore Drilling's fmal report.
This research created a decision model, using Simple Multi-Attribute Rating
Technique (SMART) that can be applied to prioritizing resource allocation of federal
fines from the BP deepwater Horizon incident in a fashion that promotes resilience
measures towards future 'oil related incidents in Louisiana coastal communities. This
research focuses specifically on the needs of St. Bernard Parish, LA (St. Bernard). The
decision analytic tool elicits the preferences and value trade-offs of experts with
significant knowledge of the issues facing St. Bernard to create a decision analytic tool.
The multi-criteria decision model establishes a priority based ordinal ranking and
preference structure of key aspects necessary for long-term resilience in St. Bernard
Parish following the BP Deepwater Horizon incident. These aspects will be delineated as
objectives and sub-objectives of an optimal response. The objectives to be included in
this study include the environment, economic stability, logistical capacity for disaster
response, and societal impacts.
The ordinal ranking will be accomplished by collecting insights from experts
obtained through an interview and questionnaire process. Experts will be selected from
the fields of public health, environmental science, academia, oil and gas, elected
10
government officials, government employees, engineers, community oriented non-profits,
and representatives from the fishing industry. At the conclusion of the interviews, a
multi-criteria value tree will be constructed. The priority of objectives of a response will
be established through a weighting process, based on the expert input.
The following research questions will be addressed:
1. What are the model variables to be addressed in this multi-criteria value model that promotes long-term resilience with consideration to future oil related events in St. Bernard Parish, LA?
2. What are the differences, values, and preferences across and among the expert's identified expert groups?
3. Can a decision model provide a useful tool and template to decision makers for promoting resilience with consideration to future oil related events?
4. What are the weights of model variables and how are they prioritized?
5. What can be learned through the development of this model?
The assumptions of this study are:
• The data collection process requires professional knowledge that is consistent in the experts' judgment.
• The experts are the best-qualified individuals locally available to respond to the questionnaires.
• The experts tend to make consistent judgments with one another.
• The diversity and scope of the educational and professional background of the experts reflect their expertise as attributed to their roles as members of their profession.
• Monetary resources resulting from penalties associated with the BP Deepwater Horizon are limited and need to be maximized to best promote resilience.
11
Dermitions:
The definitions and use of terms that are frequently referred to throughout this research are below:
Multi Attribute Decision Analysis: "an umbrella term to describe a collection of formal approaches which seek to take explicit account of multiple criteria in helping individuals or groups explore decisions that matter"(Belton and Stewart 2002).
Analytical Hierarchy Process (AHP): A form of Multi-Attribute Decision Analysis that measures alternatives in a pairwise comparison. The comparisons indicate the expert's judgment and values on prioritizing alternatives (Saaty, 1980). In terms of this research the Analytical Hierarchy Process will be used to create a prioritized list to decision makers for allocating limited resources in the event of a man-made water related disaster. The specific type of Analytical Hierarchy Process used in this research is the Simple Multi Attribute Rating Technique (SMART).
Simple Multi-Attribute Rating Technique (SMART): A technique of multi attribute decision model that assesses the utility of alternatives in a direct rating technique (von Winterfeldt, 1986).
Disaster: "A disaster is a sudden, calamitous event that seriously disrupts the functioning of a community or society and causes human, material, and economic or environmental losses that exceed the community's or society's ability to cope using its own resources" (IRCRC, 201 0). The International Red Cross and Red Crescent provide an equation to identify the relationship of the components of a disaster: Vulnerability+ Hazard/Capacity =Disaster
Disaster management: "The managerial function charged with creating the framework within which communities reduce vulnerability to hazards and cope with disasters" (FEMA framework, 2011).
Deep-water oil drilling: Oil and gas exploration at water depths greater than 156 meters or 500 feet, as defmed by the US government temporary moratorium placed on deepwater oil drilling operations in May 2010 (Department ofthe Interior, 2010).
Oil Leak: The "sudden, localised release of petroleum into the environment. In the case of major spillages, the quantities greatly exceed what the local environment is able to assimilate without resulting in damages" (CEDRE, 2011 ).
Values: The opinions of experts on what they see as the most vital aspects of a decision.
Objective: The individual components that when pieced together, comprise the decision in question (Ramanathan, 2001).
12
Sub-objective: The variables that comprise of the objectives in the decision model.
Trade-offs: A measurement of potential gain from selecting one objective at the detriment of a different objective (Keeney, 2001 ). Trade-offs in decision modeling identify expert's values.
Weight: The numerical value experts give to their perceived importance of an attribute. Weighting in Multi Criteria Decision Analysis is the tool used to assess Trade-offs (Phillips & Stock, 2003).
Resilience: The Community and Regional Resilience Institute defmes as "The capability to anticipate risk, limit impact, and bounce back rapidly through survival, adaptability, evolution, and growth in the face of turbulent change".
13
~~~~·-~- --~~·------------
CHAPTER II
LITERATURE REVIEW
2.1 Multi Criteria Decision Analysis:
The nature of decision-making is often a complex process involving interconnected
issues of which outcomes can have far reaching impacts on the experts involved.
Decision makers must approach a decision in the most logical and infor:ined fashion to
fully understand the depth of the decision and subsequent ramifications. Multi Criteria
Decision Analysis (MCDA) is a methodology that dissects a complex decision into its
most basic elements of importance. MCDA presents a tool that considers quantitative and
qualitative information in the decision.
In the process of decision model development the attributes and alternatives of the
decision are separated into objectives and sub-objectives and placed into a graphic model
known as a value tree. A decision tree is a graphical tool that aims to dissect components
of a decision in a logical fashion for decision makers to have as clear of an understanding
of outcomes from a decision as possible.
Once the attributes have been placed into the value tree, experts are interviewed and
asked to rank and rate the sub objectives in what they see as the most important to the
least important. This results in the development of an order of priority with the
objectives and sub-objectives. At the conclusion of the interviews, the researcher will
perform statistical analysis, and weights will be assigned to the objectives and sub
objectives. The sub objective and objective with the largest weight represents what the
14
experts collectively see as the most important areas to consider in the resource allocation
decision (Y oe, 2002).
A benefit of decision modeling is the determination of objective utility scores that are
used as a method to quantifY experts' values for decision makers. Objective utility scores
scientifically dissect a problem and express the values of experts in the field. This is a
helpful tool for the decision maker to have in explanation of the rational and logic behind
the decision. The decision model considers values placed on alternatives, or value trade
off's, by the identified experts. Value trade-offs can best be described as the priorities of
the experts in the creation of the model and the decision maker in the choice of an
alternative. It is the importance placed in selecting one alternative at the detriment of the
other alternatives. The decision analysis assesses the decision maker's value trade offs of
the alternatives in the decision opportunity through an objective score known as a utility
or value score (Keeney, 1982). Both utility and value scores measure the decision
maker's preferences for decision alternatives. They differentiate in that utility scores
measure the preference of an uncertain alternative by a decision maker, while value
scores measure preference in certain alternatives. The use of a value model requires the
decision maker assess their value trade-offs and willingness to accept risk (Keeney,
1982).
The utility score and the subsequent value trade-off analysis of the experts will be
assessed through a series of questionnaires, conversation with the experts, and statistical
analysis of their response to the questionnaire. A portion of the statistical analysis will
include weighting the expert responses. The weighted values will provide important
insight to the decision maker on the value that the experts from a given field rank as a
15
priority. This process allows the decision maker to gain knowledge of the basic
components of the decision, to provide the decision maker an opportunity to understand
the values of the experts, and to provide the decision maker time to consider the decision
and provide insight into the decision makers own values as applied to the event (Alemi,
2007).
The purpose of this model is to provide the decision maker(s) a prioritized list
detailing the values that are likely to be affected in the event of an anthropogenic disaster
effecting water quality (similar to that of the BP oil leak). The allocation of resources
during a disaster of this nature will ultimately be dictated by elected officials who may
not have extensive knowledge of oil related disasters. Governing officials in the United
States are elected on a broad platform and may have limited understanding of a given
problem.
Further, it is not unusual for human nature to quickly respond to a given situation
without thoroughly thinking the process through. This process indicates reactive, not
proactive decision-making, also referred to as alternative-focused thinking (Keeney,
1996). Decision-making requires the decision maker to fully understand the components
of a problem and what experts have to gain or lose by the outcome of the decision. If the
components are not fully understood by the decision maker, the decision maker values
may reflect only his or her values, and the full breadth and complexity of values may not
be displayed in the outcome of the decision (Alemi, 2007). Should this occur, it does not
mean that the decision is right or wrong but rather ill informed, carrying a potential for
undesirable outcomes. In context of this research, the complexity of the social ecological
system needs a multidisciplinary approach to address the allocation of resources
16
following an oil leak. A MCDA framework can address this need. An additional benefit
to implementing a structured decision model is that it allows the decision maker an
objective and valuable tool to reference should the decision come into question.
Two critical factors are necessary in order to confirm the value of the decision model.
The first is the process of selecting experts to participate in the ranking of the values.
The second is the need of the researcher to maintain objectiveness when interviewing the
experts (Alemi, 2007). Those chosen to participate in the interview session will be
divided into scientific/technical, industry, citizen, appointed officials, and regulatory
expert groups. The pool of experts was chosen based on their education, years of
experience, knowledge of problems facing St. Bernard Parish, and involvement in their
given field throughout the ongoing response to the BP oil disaster in the Gulf of Mexico.
They will initially be identified through the researcher's attendance at public meetings,
conferences, other outreach programs, and recommendations. In addition to involvement
in the current response efforts, professional experience and education level are
parameters that will be considered in defining and identifying experts. Those selected
will comprise a pool stemming from a diverse background in engineering, environmental
science, disaster management, oceanography, local and state government representatives,
natural resource industry managers, healthcare industry managers, fisherman, citizens,
and representatives from the tourism industry. During the interview stage, the researcher
asked the interviewee of knowledge of other persons they consider experts in their field.
Those suggested were evaluated on whether or not they qualify as experts in this
research. Those who qualified as experts were then contacted in order to gauge interest
Mitigation in disaster management focuses efforts in avoiding any unnecessary
outcomes associated with disasters and includes legislation like building codes and
proper urban planning (FEMA, 2003).
Preparedness in disaster I11anagement includes the development of a preparedness
plan to minimize the effects of a disaster and to ensure that proper infrastructure and
logistical capacity is in place or can be implemented in the event of a disaster. Response
in disaster management is the implementation of the preparedness plan. The fmal stage
in the disaster management paradigm is the recovery stage. This phase can take years to
complete depending on the severity of the disaster as well as socioeconomic and political
factors (FEMA, 2003).
Following the Deepwater Horizon oil spill, the Department of the Interior established
the Strategic Sciences Working Group (SSWG). The objective ofthe SSWG was to
identify potential economic, environmental, and social impacts of the oil spill on the Gulf
42
Coast communities. As a result of SSWG meetings, the group created impact and
management phases following an oil spill. They were the emergency phase, the
restoration phase, and the reconstruction phase (Machlis, 201 0). The proposed research
will create a list of prioritized resilience measure to be implemented in St. Bernard
Parish that promote resilience against future oil related incidents in the Gulf of Mexico.
2.5 Application of Multi Criteria Decision Analysis to Resource Allocation and Oil
Related Incidents:
The inconsistencies found in BP's Gulf of Mexico regional oil spill plan further
underscores the importance of creating a new preparedness plan for oil related operations
in the Gulf of Mexico. BP's Gulf of Mexico regional oil spill plan was heavily criticized
as being insufficient and lacking useful information. Examples of inaccuracies included
citing wildlife protection of species not found in the Gulf of Mexico (walruses and sea
lions) and thoughtless estimates of worst case scenarios (National Commission on the BP
Deep Water Horizon Oil Spill and Offshore Drilling, 2011, New York Times, 6/14/201 0;
The Economist, 6/17/201 0; BP Gulf Oil Spill Response, 2009). This disaster illustrated
the need to create a practical response plan following an oil related incident in the Gulf of
Mexico. A component of this plan should address resource allocation following the
disaster.
These scenarios present varying impact assessments and intervention points following
the BP oil spill, however do not give recommendations on priority in addressing the
issues. As of this writing SSWG has not produced further documentation beyond the
scenario building associated with the BP Deepwater Horizon incident.
43
Multi Criteria Decision Models and particularly the Simple Multi-Attribute Rating
Technique (SMART) can be a useful tool as part of a disaster preparedness plan. This
model will provide an assessment of expert values and perception as to the priorities of
various resilience measures to be implemented in St. Bernard Parish that promote
resilience against future oil related incidents in the Gulf of Mexico. The constructed
prioritized list resulting from the SMART model will be useful to decision makers as
funds released following the oil spill will require a quick and informed response that
appropriately, efficiently, and effectively manages and mitigates the disaster.
2.5.1 Analytic Network Process (ANP):
A Multi-Criteria Decision Analysis using the Analytic Network Process (ANP) was
conducted during the time that the BP Deep Water Horizon oil spill was occurring. The
assessment by Levy & Gopalakrishnan was published online July 15,2010. ANP is
considered a broad version of the AHP (Saaty, 1999). The focus of this ANP was
resource allocation during BP's Deep Water Horizon oil spill and impact on the Gulf
coast. Their assessment included four objectives: higher economic payments to coastal
business; increase use of dispersants (meaning less oil will get to the shore); berm
construction to prevent oil from getting to the shore; and rebuilding Gulf Coast wetlands,
strengthen regulations, and promote sustainability and resilience (Levy, 2010). Details of
the experts' professional affiliations and the sample size used for the model were not
given. The completed rankings from the model can be seen in the following chart:
44
Alternatives Rankings Berm construction only 0.385 Rebuilding Gulf Coast wetlands, strengthen 0.304 regulations, promote sustainability and resilience Increase use of dispersants 0.196 Higher economic payments 0.115
Levy & Gopalakrishnan research is an assessment on allocating funds during the
emergency response phase for the BP Deepwater Horizon incident and assesses four
broad measures.
2.5.2 SMART and Resource Allocation:
In comparison to the previously mentioned decision model, the research conducted in
this dissertation uses a different methodology, assess the long-term resilience of St.
Bernard Parish, LA (opposed to emergency response), and provides a more
comprehensive assessment of resilience issues facing the coastal region.
The results of this research will be used in the development of a practical tool and
resource for decision makers to reference in the face of a future oil related incident in the
Gulf of Mexico. This tool can be integrated into the development of a disaster
preparedness plan.
The primary source of data for the research project used experts' experiences
associated with the BP oil leak. The timing of the data collection is after the emergency
response and the permanent capping of the oil well that occurred in September 2010.
Due to the length of time from the oil leak and capping of the well to the time this study
was conducted, the experts and their perceptions and values of resource allocation
45
following the oil incident may have changed values since the emergency response phase.
46
CHAPTER Ill
RESEARCH METHODOLOGY
The created decision model serves as an approach towards resource allocation using
financial penalties associated with the Deepwater Horizon oil leak and promotes
preparedness in St. Bernard in the event of a future oil related incident in the Gulf of
Mexico. The model creates a prioritized list of response steps to take following an
incident that will be useful to decision makers in the timely distribution of limited
resources for recovery efforts. It is important to note that this model is intended to
provide recommendations and guidance to decision makers in the face of future oil
related events in the Gulf of Mexico. The resource allocation model developed integrates
key experts perceptions and values. These values were assessed through an interview
process that asked experts from all expert groups to rank a number of pre-identified
objectives and sub-objectives as to their importance in relation to response activities.
Upon completion of the interview sessions, all results were statistically analyzed using
the Kruskal-Wallis H-test and Kendall's Coefficient of Concordance.
The creation ofthis model intends to assess the value trade-offs of the experts in an
objective manner. Each expert was assured anonymity throughout the completion of this
research in order to best obtain the value trade-offs of experts without restraint.
3.1 Development of the Model
The initial development and framing of the decision model began with a thorough
literature review of published articles involving oil spills resulting from deepwater oil
47
drilling incidents, decision modeling, Social Ecological Systems (SES' s ), and disaster
management. The literature review provided the researcher insight into the areas needed
for development of objectives and sub-objectives for the development of the decision
model.
The selection of the objectives and sub-objectives considered the criteria outlined in
Keeney & Raffia's 1992 book, Decisions with Multiple Objects. It is stated that the
decision objectives and sub-objectives should consider completeness, operational
function, absence of redundancy, succinctness and decomposability. Decomposability
refers to the ability to analyze an option by separating it into smaller components
(Hankins & Fos, 1989). The objectives support the overarching goal of the decision to be
made. This research will contain four objectives, Environment, Logistical Capacity for
Disaster Response, Regional Economic Stability, and Societal impacts. Sub-objectives
provide support and detail to each objective.
The model variables to be implemented into the SMART questionnaire for this
research were identified through use of the Delphi method. The Delphi method is a
process by which experts reach consensus answer a questionnaire to rate variables using
the Likert scale. This study used a modified Delphi method. Rather than giving an open
ended questionnaire in the first round, experts were asked to evaluate structured
statements developed through a focus group organized by the researcher. The objective
and sub-objectives results from the focus group can be found in Table 3-1. Following the
expert response, the questionnaires are gathered by the researcher and analyzed. The
researcher averages the responses given and calculates the standard deviation of each
variable. The results from all experts' responses from the first round are entered on the
48
questionnaire and given back to the experts for review. The experts assess the results
from their peers andre-rate each variable. This process is continued until consensus is
established. This study uses the rating of 3.5 on the Likert scale and a standard deviation
ofless than one as consensus. Below are the structured statements (sub-objectives)
identified by focus group to be rated in round one of modified Delphi method.
Environment:
Removal and disposal of residual oil: Allocate resources to cleaning oil that should surface, wash-up to shore, or invest in technology that best addresses underwater oil plumes.
Construct Oyster Reefs: Oyster reefs would promote shoreline stability and contribute to the ecosystem of Lake Borgne, St. Bernard Parish.
Cypress Swamp Restoration: Allocate funds to reroute the use of wastewater and freshwater to flow into the cypress swamps to mitigate the impacts of saltwater intrusion and promote cypress swamp restoration.
Prevention of Herbivory: Trapping of invasive species, primarily nutria, to reduce their impact on wetlands, therefore providing a healthier and more resilient wetland system. We have found that shootings are much more successful than trapping
Restoration of Barrier Islands: Infusion of sediment to counter the landmass lost to the barrier islands associated with erosion.
River Diversions: Use available funds for river diversions (Canarvon and Violet) in St. Bernard Parish to promote the introduction of freshwater to the marshes in St.· Bernard
Parish.
Sustainability of fisheries: Conduct studies quantifying the impact of the oil leak on fish populations and analyzing the long-term impact.
Logistical Capacity for Disaster for Response:
Basic Equipment and Necessities for residual oil clean-up: Boom, boats, dispersants, Personal Protective Equipment, food and shelter for workers, fuel.
49
Environmental Health Education and Training: HAZWOPER training, and other related training for programs to those needing to adjust their occupation to relevant and available work following an oil leak.
Establish monitoring system: Environmental Sensitivity Index, use of GPS, and methods that allow ease of reporting of oil sightings.
Information Sharing: Develop and implement systems for data integration, synthesis, sharing and dissemination.
Streamline Recovery Fund Procedure: Provide additional resources to community, particularly workers who can describe the paperwork that is required within applications.
Parish Staff I Parish Administrative capacity: Provide funds for additional St. Bernard Parish staff and administration to handle increase administrative needs associated with the BP Deepwater Horizon incident.
Funding for increased community meetings and programs: Promote understanding of all aspects of recovery and restoration of a disaster between expert groups.
Regional Economic Stability:
Economic Outreach Programs: Create television commercials and magazine adds to promote tourism.
Aquaculture Studies: Finance for independent laboratories to examine extent of contamination and bioabsorptionlbioaccumulation of aquaculture.
Restaurant and Hotel Subsidies: Subsidies for restaurant and hotel owners and workers to account for 100% of 5-year average profit prior to disaster.
Stipend for business adjustment to new environmental regulation: Provide money to small business that fund updating their business to meet new environmental regulation resulting from oil spill.
Loans for small business development: Increase availability of micro loans to individuals for small business use and/or start-up from affected communities.
Government funding: Provide local governments with funds to compensate reduced tax revenue as a result of reduced industry output.
so
Societal Impacts:
Community education outreach: Provide outreach sessions to communities that provide scientific backing on the health outcomes and effects of the oil and dispersants.
Bolster Educational Programs: Promote education and literacy in rural and urban regions.
Public health staffing and databases: Increase states resources for surveillance of adverse health effects associated with oil and dispersants.
Improve capacity of public health programs: Increase funds for social programs such as food stamps, school lunches, immunizations, and maternal health for effected populations.
Increase mental health care capacity: Develop resources and providers to deal with mental health problems, like Post Traumatic Stress Disorder (PTSD), to communities affected by the spill.
Population and demographic studies: Establish baseline data following the oil disaster to help determine future resource allocation.
Language translation: Provide language translation services to those in need.
3.2 Structuring the Decision Model
Consensus of model variables for the decision model was reached at the completion
of the second round of the Delphi method. The decision question was separated in a
decision tree. At the head of the decision tree was the title of the prioritized list of
resource allocation following an oil leak that promotes the long-term resilience of St.
Bernard Parish, LA. The problem was separated into four branches with the objective
titles. Each objective had corresponding branches for each sub-objective. The decision
tree for this research is found in Chapter 4, Figure 4-1
The objective categories and initial sub-objectives for the model were established
through the literature review in order to allow the model framework to have an emphasis
51
on social ecological systems. The researcher conducted a focus group session to refme
and modify the objectives and sub-objectives prior to the start of the Delphi rounds.
3.3 Survey Development:
The model variables identified through the Delphi method were used to develop a
survey that could used in order to determine the weights of the objectives and sub
objectives in the decision tree. The methodology used in development of the survey was
the Simple Multiattribute Rating Technique. The survey can be found in Appendix A.
3.3.1 Internal Review Board Approval:
The completed survey was submitted to the Internal Review Board (the IRB) at
Tulane University. The survey received expedited approval from the IRB on September
20, 2011 to conduct the proposed study with thirty participants. The request to omit
written consent was denied by the IRB. Prior to each interview and questionnaire session
the researcher discussed the goals of the study. The researcher answered any
questionnaires the participant had that pertained to the nature of the study goals as well as
questions regarding the collection of the data, written consent, and maintaining the
anonymity of all participants. The IRB approval can be found as Appendix B
3.3.2 Selection of Experts:
In order to make this decision model practical, effective, and useful, the data collected
for this research was to be of high quality. This was accomplished through the selection
of qualified experts. Those chosen to participate in the interview session were divided
into expe1i groups of scientific/technical, business representatives, government
52
representatives, and community based groups. For the purpose of weighting and
statistical analysis, each group had a minimum of six expert representatives from each
expert group. Participants were selected based on work experience in their given field
and direct working experience with the Deep Water Horizon oil incident. As an incident
of this nature can have detrimental effects on all facets of society, the research method
intends to capture the viewpoints and values of various stakeholder groups. An attempt
was made to gather equal number of representatives from each group.
The recruitment phase of the study was difficult and time consuming. The primary
mechanism used for the Delphi method was contacting the researchers' network of
professionals to gauge interest in participating in the study. Initial outreach for
participation in the Delphi method had varied success. A total of twenty-seven people
were contacted before the appropriate fit between the experts' background and experience
and a willingness to participate was found. The experts selected had a minimum of five
years experience in a resilience related field, extensive knowledge of issues facing
Southern Louisiana, and were impacted and/or had experience with the recovery process
following the BP Deepwater Horizon oil leak. The initial two rounds of this survey had
fifteen participants with a wide range of expertise.
Recruitment for participants in the SMART survey followed a similar methodology
of recruitment as the Delphi portion of the study. The participants of the Delphi method
were asked if they were willing to participate in the resource allocation instrument. Nine
of the fifteen participants agreed. Six participants either did not respond to the request or
replied that their schedules would not allow for further participation. Nine people
participated in the questionnaires and interviews. At the conclusion of each session, the
53
study participant was asked to recommend two to three people to participant for this
study. The recommended participants were contacted about their interest in participating
in the study. This method proved to be useful in fmding qualified participants.
Research on community groups that are active in St. Bernard Parish provided an
additional list of potential study participants. The researcher contacted the community
organizations to explain the study and gauge the level of interest in participating. This
resulted in varied success. Several groups were very interested and willing to participate,
while some groups never returned calls or emails. Selected participants in this portion of
the study had diverse backgrounds and a minimum of five years experience in the fields
of environment, fisheries, oil and gas, elected officials, government employees, biology,
chemistry, health sciences, non-profit organizations, community advocacy, and
economics and experience with the BP Deepwater Horizon oil leak. The SMART study
participants were placed in four expert groups. As previously mentioned, data was
collected in two phases: the Delphi method and the SMART interview and questionnaire
session. Data collection for the Delhi portion of the study took approximately eight
weeks from July 12, 2011 to September 14, 2011. The data collection phase for the
SMART interview and questionnaire took approximately seven weeks from September
20, 2011 to November 7, 2011. In general, study participants were met on an individual
basis. The length of the interview and questionnaire session ranged from forty-five
minutes to two hours. The data collection (Delphi and interview and questionnaire)
process was consistent with all participants.
54
Study participants had a diverse background and represented the following entities:
Tulane University School of Public Health
Sewerage and Water Board of New Orleans
City ofNew Orleans Department of Homeland Security
Tulane University Environmental Law Clinic
Southeast Louisiana Association of Contingency Planners
St. Bernard Economic Development Foundation
St. Bernard Parish Government
AVODAH
Lake Pontchartrain Basin Foundation
United States Army Corps of Engineers
Louisiana State University School of Coast and Environment
St. Bernard Community Center
Gulf Environment Associates
American Red Cross
In addition to listed firms and organizations above, four consulting firms that assist
the seafood industry and oil and gas requested their organization name not be included in
the study.
3.4 Data Collection:
As previously mentioned, data was collected in two phases: the Delphi method and
the SMART interview and questionnaire session. Data collection for the Delhi portion of
the study required approximately eight weeks from July 12,2011 to September 14, 2011.
The data collection phase for the SMART interview and questionnaire required
55
approximately seven weeks from September 20, 2011 to November 7, 2011. In general,
study participants were met on an individual basis. The length of the interview and
questionnaire session ranged from forty-five minutes to two hours. The data collection
(Delphi and interview and questionnaire) process was consistent with all participants.
The interviews and questionnaires were primarily conducted at the offices of the selected
participants. However, several study participants selected to conduct the sessions at a
cafe or restaurant of their choice.
3.4.1 Weighting the Variables:
The interview and questionnaire stage of the study asked experts representing four
expert groups to rank the objectives according to importance in response to resource
allocation following a large scale oil release event to promote the long-term resilience of
St. Bernard Parish, LA. The SMART methodology of variable weighting has been used
in multiple fields and established as a valid form of decision modeling. The initial
assumption of the ranking process is that all objectives are in an equal and worst state.
The objective perceived as the most important by the expert will be given the rank of one;
the objective perceived second most important will be given a rank of two. This process
will continue for all objectives to complete the ranking process. At the completion of the
initial ranking, the experts will then be asked to rank the objectives in a fashion that
establishes magnitude between each objective (Fos & Zuniga, 1999). This weighting
process is referred to as direct ranking. For example, if the economic stability objective is
ranked lowest in the first portion of the interview, it will be given a value of 10. If the
second least important objective is perceived by the experts as being 10 times more
important the objective will be given a ranking of 100. All objectives will be ranked
56
against the preceding objective. Following the ranking of importance, the weights will be
normalized by the equation described in Fos and Zuniga, 1999 below:
w;= normalized weight
Wi' =individual rating by expert
Wi' Wi=---n
.Lwi' i=l
2: Wi '= Sum of all individual ratings by experts
n = the number of variables
Following the initial steps of ranking the objectives, the same process was completed
for sub objectives within each objective category. After calculation of the raw values,
model variables of all participants were averaged before attaining the trade-off score.
The trade-off score were attained by multiplying the normalized weight of each sub
objective by the normalized score of its relevant objective (Fos and Zuniga, 1999 &
Alemi, 2007). The equation to accomplish this is below:
n
V. = ~w.v. J LJ l l
i=l
Yj= The overall score for the j alternative
W F Each individual variable normalized weight
Vi = Each individual variable value
n =The number ofvariables
57
The result of this method creates the prioritized list of resource allocation following a
large scale oil related event promoting long-term resilience in St. Bernard Parish, LA.
3.5 Data Analysis:
Kendall's Coefficient for Concordance and the Kruskal-Wallis H-test were used in
this research to determine statistical relationship between the expert groups. The results
of the statistical analysis will provide insight to the views and values found among and
within study participants.
3.5.1 Kendall's Coefficient for Concordance:
Kendall's Coefficient for Concordance is a statistical analysis that measures the
strength of agreement in a given situation (Landis & Koch, 1977). This form of analysis
is useful in identifying correlations in prioritizing objects according to values, judgment,
and perceived importance (Jirapongsuwan, 2008). Kendall's Coefficient for Concordance
has a range of 0 to 1. A score of 0 represents no agreement, while a score of 1 equals
total agreement among study participants (Robinson, 1957). The null hypothesis in this
research will support no correlation between the rankings of experts, while the alternative
hypothesis will support a correlation between experts.
Hypothesis 1: There is a correlation between all expert participants in variable ranking
Hypothesis 2: There is a correlation within the Government Representative expert group in variable ranking
Hypothesis 3: There is a correlation within the Impacted Business expert group in variable ranking
Hypothesis 4: There is a correlation within the Community Based expert group in variable ranking
58
Hypothesis 5: There is a correlation within the Science and Technical expert group in variable ranking
Hypothesis 6: There is a correlation within the experts selected to participated in the Delphi method (defining model variables) and the SMART questionnaire
Hypothesis 7: There is a correlation among the experts who participated in the SMART questionnaire
3.5.2 Kruskal-Wallis H-test:
The Kruskal-Wallis H-test is accepted as the non parametric version of the parametric
One-Way Analysis of Variance and serves as a form of analysis that statistically assesses
the difference between independent groupings (Corder & Foreman, 2009). By using the
ranks, opposed to the raw data, the Kruskal-Wallis H-test will be used to analyze the
associations of ranking of model variables from different grouping of study participants.
The null hypothesis identifies no difference in the median value between the identified
expert groups, and the alternative hypothesis identifies a difference in median values for
between the expert groups. In order to reject the null hypothesis the H-value must be
greater than the critical value (Siegal &Castellan, 1988). The H-value is obtained from
the Chi-square table and using degrees of freedom (k-1), (Jirapongsuwan, 2008; Siegal
&Castellan, 1988). The following hypotheses were tested using the Kruskal-Wallis H-
test in this study.
Hypothesis 8: There is an agreement in ranking of model variables in the four expert groups
Hypothesis 9: There is agreement in ranking of model variables between the expert participants (those who completed the Delphi method and the SMART questionnaire) and the experts (those who completed the SMART questionnaire).
59
CHAPTER IV
RESEARCH RESULTS
This chapter will provide results from the study, including the response rate, the
response from the Delphi portion of the study, results from the SMART questionnaire,
and the results from Kendall's Coefficient of Concordance and the Kruskal-Wallis H test.
The model variables will then be sorted from highest weight to the lowest weight,
creating the prioritized list of resource allocation. The cumulative percent model was
utilized to provide additional organization to the list and to identify which model
variables should be included in the face of limited financial resources.
4.1 Response Rate
The initial phase of this research was to identify pertinent variables for the study.
Fifteen experts were recruited to defme the decision model variables through use of the
Delphi method. Consensus on twenty-nine model variables was attained at the
completion of the second round. Attrition was not an issue during this portion of data
collection. The study anticipated using the fifteen experts who participated in the Delphi
method to complete the SMART interview and questionnaire phase of the study.
However six participants chose not to complete the study. Three of the six participants
who declined to complete the SMART interview and questionnaire expressed lack of
time to continue participation in the study. The remaining three participants failed to
respond to the researcher's repeated attempts to schedule time for the SMART interview
and questionnaire.
60
Thirty experts participated in the SMART interview and questionnaire phase of this
study. The participants included the nine experts who expressed a willingness to
continue participation following the Delphi portion of the study.
4.2 Delphi Method Results:
Two rounds of the Delphi were completed before consensus was reached on model
variables for the study.
4.2.1 Round 1 Results:
The initial round of the study asked fifteen experts to rate variables to the importance
of resource allocation that promotes the long term resilience of St. Bernard Parish
following the BP Deepwater Horizon incident. The established inclusion criteria for the
Delphi method in this study were variables that rated both with a mean of 3.5 or above
and a standard deviation less than one. This level was established from the literature
review and revised due to the anticipated diverse views and interests among study
participants. The variables were categorized into four objectives: Environment,
Logistical capacity for disaster response, Economic Stability, and Societal impacts. First
round results can be found in Table 4-1.
Table 4-1: Round 1 results ofthe Delphi method
Environmental Sub-objectives
Variables Mean
Removal and Disposal of Residual Oil 4.16
Construction of Oyster Reefs 4.20
61
Standard Deviation
0.937
0.861
River Diversions 4.46 0.967
Cypress Swamp Restoration 4.73 0.457
Herbivory Prevention 3.86 0.862
Restoration of Barrier Islands 4.33 0.975
Examine Sustainability ofFisheries 4.40 0.828
Additional Variables Suggested by - -
Experts
Funds for Research Impacts of Rising Energy Costs
Funds for Researching Impact of Climate Change
Monitoring natural attenuation of residual oil
Logistical Capacity for Disaster Response
Variables Mean Standard Deviation
Basic Equipment and Necessities for Residual Oil Clean-Up 3.86 1.060
Education and Training 4.13 0.833
Establish Monitoring System 4.06 0.883
Information Sharing 4.33 0.816
Streamline Recovery Fund Procedure 4.46 0.767
Funds for Increased Community Meetings and Programs 4.00 0.925
Removal and Disposal of Residual Oil Cypress Swamp Restoration Construction of Oyster Reefs River Diversions Restoration of Barrier Islands Herbivory Prevention Examine Sustainability of Fisheries Researching Impacts of Climate Change Monitoring Natural Attenuation of Residual Oil
Basic Equipment for Residual Oil Clean Up Education and Training Establish Monitoring System Information Sharing Streamline Recovery Fund Procedure Funds for Increased Community Meetings/Programs Parish Administrative Capacity
Funding for Government Outreach Programs to Promote Tourism/Fisheries Aquaculture Studies Stipend for Business Adjustment to New Enviro Regulation Loans for Small Business Development
Community Outreach and Support Programs 2.450 0.484 Bolster Education Programs 2.160 0.540 Public Health Staffmg and Surveillance Database 2.350 0.504 Improve Capacity for Public Health Programs 6.590 0.086 Increase Mental Health Capacity 4.060 0.254 Population and Demographic Studies 1.203 0.752 Language Translation 0.004 1.000 Improve Medical Services 0.566 0.904
Hypothesis 9: There is agreement in ranking model variables between those that defmed the model variables and those that did not.
The Kruskal-Wallis H-test was used to analyze the level of agreement between the
study participants who defmed model variables. The group that defined model variables
participated in the two rounds of the Delphi method and completed the SMART
interview and questionnaire session. The significance level was set at 0.05. The
comparison of median ranked priorities of objectives and sub-objectives between the two
groups indicate statistical significance for two sub-objectives within the environment
objective. Those are removal and disposal of residual oil and river diversions. The
results indicate acceptance of the null hypothesis of no difference in median ranking
between the two groups for thirty-one of the thirty-three model variables.
Table 4-11: Kruskal-Wallis H-test ofvariables for experts' agreement among two
groups, those that identified model variables and those who did not ( df = 1):
Objective Environmental
Chi-Square p-value
Logistical Capacity for Disaster Response Regional Economic Stability Societal Impacts
81
0.090 0.765 0.920
0.001 0.442
0.338
0.981 0.506
- - -------------
Removal and Disposal of Residual Oil 4.630 0.031 * Cypress Swamp Restoration 0.557 0.456 Construction of Oyster Reefs 0.042 0.837 River Diversions 9.087 0.003* Restoration of Barrier Islands 0.053 0.819 Herbivory Prevention 0.844 0.358 Examine Sustainability ofFisheries 1.408 0.235 Researching Impacts of Climate Change 0.089 0.765 Monitoring Natural Attenuation of Residual Oil 1.657 0.198
Basic Equipment for Residual Oil Clean Up 1.113 0.292 Education and Training 0.360 0.548 Establish Monitoring System 0.089 0.765 Information Sharing 0.237 0.627 Streamline Recovery Fund Procedure 0.511 0.475 Funds for Increased Community Meetings/Programs 0.378 0.538 Parish Administrative Capacity 0.009 0.925
Funding for Government 0.124 0.725 Outreach Programs to Promote Tourism/Fisheries 0.035 0.852 Aquaculture Studies 2.260 0.133 Stipend for Business Adjustment to New Enviro Regulation 0.199 0.656 Loans for Small Business Development 0.002 0.961
Community Outreach and Support Programs 0.967 0.325 Bolster Education Programs 0.432 0.511 Public Health Staffing and Surveillance Database 0.089 0.765 Improve Capacity for Public Health Programs 0.541 0.462 Increase Mental Health Capacity 0.174 0.677 Population and Demographic Studies 0.485 0.486 Language Translation 0.984 0.321 Improve Medical Services 1.860 0.173
4.4.3 Hypothesis Conclusion and Summary:
Significance was established for twenty-nine of thirty-five variables using Kendall's
Coefficient of Concordance (W). Significance establishes agreement among the experts
and their grouping. Levels of agreement among experts were stronger within their expert
82
------- ------------ .. -- -------- ----
group opposed to the collective group of experts. Results between the experts that
identified model variables and those that did not, found higher levels of agreement within
the group that identified the model variables, however those that did not identifY model
variables results were more consistent among ranking variables. Accepting the Kruskal-
Wallis H-test null hypothesis indicates agreement among the median ranking of variables
by experts in the four designated groups and between the group that identified model
variables and participated in the SMART questionnaire and those who only participated
in the SMART questionnaire. There was little agreement among the four identified
expert group in median ranks of the environmental model variables. The outlying expert
group in the median rankings for those environmental variables with p-values below the
significance level of 0.05 was identified as the business representative group.
Summarized result of the hypothesis testing used in this research can be found in
Table 4-12.
Table 4-12:
Hypothesis Statistics Result
1. There is correlation among all expert participants in Kendall RejectHo variable ranking.
2. There is a correlation within the Government Kendall Reject Ho Representatives expert group in variable rankings.
3. There is a correlation within the Impacted Business Kendall RejectHo"' expert group
4. There is a correlation within the Community Based Kendall Reject Ho expert group
5. There is correlation within the Science and Technical Kendall Reject Ho expert group
83
6. There is correlation within the expert group that defined model variables.
7. There is a correlation within the experts not involved in defining model variables
8. There is an agreement in ranking of model variables in the four expert groups
9. There is agreement in ranking model variables between those that defmed the model variables and those that did not
* ** *** ****
Rejected for four of five variables Rejected for three of five variables Accepted for twenty-five of thirty three variables Accepted for thirty-one of thirty three variables
Kendall RejectHo
Kendall RejectHo
Kruskal-Wallis AcceptH0
Kruskal-Wallis AcceptH0
The results of the statistical hypothesis testing indicate an agreement in ranking of
model variables for the resource allocation decision tool, indicating that the identified
model variables are valid for the model. The validity of model variables answers the third
research question: Can a decision model provide a useful tool and template to
decision makers for promoting resilience with consideration to future oil related
events? The decision model can provide a useful tool and template for policy makers.
4.5 Cumulative Percent and Prioritized List of Resource Allocation
The trade-off scores were placed in order of highest to lowest, with highest trade-
off score indicating the highest priority and the lowest trade-off score represents the
lowest priority. The ordering of trade-off scores provides a prioritized list of resilience
measures for addressing issues still relevant to the BP Deepwater Horizon incident as
well as future oil related incidents for St. Bernard Parish.
84
The cumulative percent model was implemented to optimize utility and address
scenarios of limited (and unknown) quantities offmancial resources associated with
penalties associated with the Deepwater Horizon. Established in past research (Mack,
2008, Jirapongsuwan, 2008, Zornes, 2007), the cumulative percent model has been used
to establish a minimum, expanded, and optimal list with cut-off points based on
percentage levels. The definition ofthe established lists can be found in Table 4-13.
Table 4-13:
List Cumulative Percent
Minimum 0-70%
Expanded 0-80%
Optimal 0-90%
The available funds resulting from BP penalties to be utilized for resilience measures in
St. Bernard Parish will be limited. Pending the outcome of litigation, policy makers can
utilize the appropriate list (minimum, expanded, or optimal) to implement according to
available funds.
4.5.1 Results for all expert groups
The model variables were sorted by trade-off scores from highest to lowest, thus creating
the prioritized list of resource allocation. This answers research question number four:
What are the weights of model variables and how are they prioritized? Trade-off
scores were then used in a cumulative percent list as a tool to assist policy makers in a cut
85
off mark for resource allocation when a minimum amount of funding is available. The
prioritized list of model variables and the resulting cumulative percent list can be found
in Table 4-14.
Table 4-14
Variables Minimum List Loans for Small Business Development Cypress Swamp Restoration Removal and Disposal of Residual Oil Restoration of Barrier Islands Aquaculture Studies River Diversions Outreach Programs to Promote Tourism/Fisheries Increase Mental Health Capacity Funding for Government Stipend for Business Adjustment to New Enviro Regulation Improve Capacity for Public Health Programs Construction of Oyster Reefs Improve Medical Services Expanded List Community Outreach and Support Programs Examine Sustainability of Fisheries Information Sharing Optimal List Public Health Staffing and Surveillance Database Monitoring Natural Attenuation of Residual Oil Basic Equipment for Residual Oil Clean Up Establish Monitoring System Bolster Education Programs Low Ranking Variables Streamline Recovery Fund Procedure Researching Impacts of Climate Change Education and Training Herbivory Prevention Parish Administrative Capacity Population and Demographic Studies Funds for Increased Community Meetings/Programs Language Translation
The trade-off scores attained through expert input prioritize the model variables. The
list of variables has been separated into a minimum list, optimal list, and an expanded list.
The minimum list consists of all variables with a cumulative percent of less than or equal
to 70%. However this research for the minimum list is extended to 71.97% and will
include the community outreach and support programs as an exception. The expanded
list consists ofthose variables found with a cumulative percent within 80% of model
variables. The optimal list includes those rankings that account for 90% of the
cumulative percent list. This list contains twenty-two of twenty-nine variables, including
the streamline recovery funds procedure. Variables with a cumulative percent greater
than 90% do not meet the inclusion criteria for the optimal list and were not considered a
priority by the expert. The seven variables were researching the impacts of climate
change, education and training, herbivory prevention, parish administrative capacity,
population and demographic studies, funds for increased community meetings and
programs, and language translation. Monetary resources available for projects following
an incident like the deepwater horizon are significant but far from unlimited. Separating
the list into the minimum, optimal, and expanded list will be useful for policy makers to
address those variables with highest priority (weight) as indicated by study participants,
with limited financial resources.
Model variables under the economic stability and environment accounted for ten of
the fourteen variables within the 70% cumulative percent inclusion criteria for the
minimum list. Comments received by the researcher during the interview portion of the
SMART questionnaire indicated concern about the lack of employment diversification
and opportunities in St. Bernard and the need for innovative strategies to bring jobs
87
and/or create jobs in St. Bernard as a solution. Coastal erosions and hurricane protection
in St. Bernard was a second issue of concern raised in the interview sessions.
The societal impact objective had three variables found in the minimal list.
Increasing mental health capacity had the highest trade-off score of 0.0404, followed by
the need to increase public health capacity (0.0357) and improve medical services
(0.0340).
The variables in the logistical capacity for disaster response were not ranked high as a
group. No variables met inclusion criteria for the minimum list. The highest trade-off
score elicited from study participants was information sharing (0.0304), and this one
along with establish monitoring system (0.0215) and streamline recovery fund procedure
(0.0203) meet the inclusion criteria for the optimal list. The remaining variables were
above the 90% cumulative percent threshold for the optimal list.
4.5.2 Results for Expert Groups
In addition to the statistical analysis, the trade-off scores and cumulative percent
model were used to assess the four identified expert groups. Though these individual
lists are not considered in creating the prioritized list of resource allocation, they do
provide further analysis on the differences between expert groups. Their results are
found in Table 4-15 through Table 4-18.
The government representative expert group optimal list consisted of four of five sub
objectives found in the economic stability objective. They included loans for small
business development (0.1368), stipend for business adjustment to new environmental
regulation (0.0786), outreach programs to promote tourism and fisheries (0.0462), and
88
funding for government services (0.0444). The researcher anticipated this expert group
would rank the economic stability variables highest. This hypothesis was anticipated due
to the continued recovery from the economic downturn in 2008. The number of
environmental sub-objectives included in the optimal list was low. Many ofthe
environmentally related sub-objectives were found towards the lower end of the list.
Funds for climate change, sustainability of the fisheries, monitoring of the natural
attenuation of oil, and herbivory prevention were all ranked in the lowest variables.
Three variables were selected, including restoration of barrier islands (0.1124), cypress
swamp restoration (0.0780), and river diversions (0.0534).
There were four sub-objectives from the societal impacts objectives. The highest
ranked ofthe four was the improvement of medical services (0.0509), followed by
increase in mental health capacity (0.0378), public health staffmg and surveillance
(0.0299), and improve capacity for public health programs (0.0222). The logistical
capacity for disaster response did not rank well among the science and technical expert
groups. Only one variable, streamline recovery fund procedure (0.0376), was included in
the expanded list with a cumulative percent of76.16% ofmodel variables. Parish
administrative capacity had a cumulative percent of89.12% of model variables, placing it
as the only variable within the objective category to be included in the optimal list, but
not the expanded list. The remaining five model variables had a cumulative percent of
90% or above. The government representatives who participated in this research did not
consider those variables important. The trade-off scores and cumulative percent for
model variables can be found in Table 4-15.
89
Table 4-15:
Individual expert groups preference on resource allocation to promote the long-term resilience of St. Bernard Parish.
Government Representatives:
Trade-Off Cumulative Variable Loans for Small Business Development Restoration of Barrier Islands Community Outreach and Support Programs Stipend for Business Adjustment to New Enviro Regulation Cypress Swamp Restoration River Diversions Improve Medical Services Outreach Programs to Promote Tourism/Fisheries Funding for Government Increase Mental Health Capacity Streamline Recovery Fund Procedure Public Health Staffmg and Surveillance Database Improve Capacity for Public Health Programs Removal and Disposal of Residual Oil Population and Demographic Studies Bolster Education Programs Parish Administrative Capacity Aquaculture Studies Monitoring Natural Attenuation of Residual Oil Examine Sustainability of Fisheries Information Sharing Herbivory Prevention Establish Monitoring System Researching Impacts of Climate Change Education and Training Construction of Oyster Reefs Basic Equipment for Residual Oil Clean Up Language Translation Funds for Increased Community Meetings/Programs
The business representative expert group had the highest level of agreement among
all expert groups for the economic stability sub-objectives. The level of cohesiveness can
be seen in the high level of prioritization of economic stability variables ranked by this
group. Four offive economic variables, loans for small business development (0.2437),
stipend for business adjustment to new environmental regulations (0.1044), outreach
programs to promote tourism and fisheries (0.1025), and aquaculture studies (0.053),
were not only included in the cumulative percent minimum list but ranked as the top four
prioritized variables by this expert group. These four variables account for a cumulative
percent for 50.36% of model variables. A large discrepancy was shown between the four
economic stability variables found in the minimum list and the remaining fifth economic
stability variable. Funding government services had a cumulative percent of92.38%
among model variables. The low level of prioritization of this variable excludes it from
any list considered in this research by the business expert group. There is a 2.5 fold
difference between the top ranked variable to the second highest ranked. The researcher
noted in the interview sessions with members of this expert group as strong preference in
the loans for small business development and innovation as critical to the survival of St.
Bernard Parish.
Following the high prioritization of economic stability variables, the environmental
and societal variables were ranked high. The environmental variables of cypress swamp
restoration (0.0364), sustainability of fisheries (0.033), and construction of oyster reefs
91
(0.0296) were all within the 70% of the cumulative percent of model variables and
included in this expert groups expanded list. The variables of increasing mental health
capacity (0.0462) improve capacity for public health programs (0.0346) and improve
medical services (0.0321) fell within the inclusion criteria for the minimum list.
Similar to the government representative expert group, the business representative
expert group trade-off scores for the logistical capacity were identified to have low
priority among model variables. The information sharing variable (0.0270) was the
highest ranked variable from that objective and would be included in an expanded list for
this group. Two variables from this group would be included in the optimal list. The
remaining four variables fall below the inclusion criteria for the optimal list. To further
note the strong preference the business expert group had towards the economic stability
variables, the four highest ranked variables (all under the economic stability objective)
accounted for 50.36% of the cumulative percent for model variables while six variables
from the environmental and societal objective account for the next 20% that comprise the
minimum list for this expert group. The value-trade off score and cumulative percent for
the business representative expert group is found in Table 4-16.
Table 4-16:
Variable Loans for Small Business Development Stipend for Business Adjustment to New Enviro Regulation Outreach Programs to Promote Tourism/Fisheries Aquaculture Studies Increase Mental Health Capacity Cypress Swamp Restoration Improve Capacity for Public Health Programs Examine Sustainability ofFisheries
Improve Medical Services Construction of Oyster Reefs Removal and Disposal of Residual Oil Community Outreach and Support Programs Information Sharing Basic Equipment for Residual Oil Clean Up Restoration of Barrier Islands Bolster Education Programs Education and Training Public Health Staffing and Surveillance Database Funding for Government Monitoring Natural Attenuation of Residual Oil Establish Monitoring System River Diversions Streamline Recovery Fund Procedure Population and Demographic Studies Language Translation Herbivory Prevention Researching Impacts of Climate Change
The science and technical group ranking of model variables included more variables
from the four objectives than the government representatives or business representative
expert groups. Seven out of nine environmental variables were ranked in the expanded
list. The highest ranked environmental variables included the highest ranked variable of
removal and disposal of residual oil (0.2254), followed by monitoring the natural
attenuation of residual oil (0.0578), restoration of barrier islands (0.0517), funds to
examine the sustainability of the fisheries (0.0390), cypress swamp restoration (0.0379).
With the exception of the construction of oyster reefs variable (0.0173), the remainder of
the environmental variables are within 90% of the cumulative percent of model variables.
The highest ranked variable, removal and disposal of residual oil was ranked over three
times more important than the second ranked variable. The science and technical expert
93
group did not prioritize the societal impact variables high, as two of eight variables would
be included in the expanded list. However, increased mental health capacity variable
(0.068) ranked as the second most important variable on the list. Community outreach
and support programs (0.0357) ranked within 70% of the cumulative percent for model
variables. Establish monitoring systems (0.0354) and information sharing (0.0336) were
the two variables from the logistical capacity for disaster response objective that meet the
inclusion criteria for the minimum list.
Despite aquaculture studies (0.0608) being the third highest ranked variable on the
list, the ranking of economic variables were considerably lower by this expert group in
comparison to other expert groups. Funding for government services (0.0444) was the
second highest ranked of the economic variables. These two variables were ranked
within the 70% cumulative percent for inclusion in the minimum list. Loans for small
business development (0.0282) was ranked substantially lower by this expert group than
the other three, though would still be included in the expanded list of priority variables
for resource allocation. Outreach efforts for promotion of fisheries and tourism (0.0240)
was ranked within the optimal list criteria. Funds for adjustment to new environmental
regulation (0.0114) rank was not sufficient to be included in the optimal list. Trade-off
scores and cumulative percent results for the science and technical group are found in
Table 4-17.
Table 4-17:
Variable Removal and Disposal of Residual Oil Increase Mental Health Capacity Aquaculture Studies
94
Trade-Off Cumulative Score
0.2254 0.0680 0.0608
Percent 22.54 29.34 35.42
Monitoring Natural Attenuation of Residual Oil Restoration of Barrier Islands Funding for Government Examine Sustainability of Fisheries Cypress Swamp Restoration River Diversions Community Outreach and Support Programs Establish Monitoring System Information Sharing Researching Impacts of Climate Change Loans for Small Business Development Basic Equipment for Residual Oil Clean Up Outreach Programs to Promote Tourism/Fisheries Public Health Staffmg and Surveillance Database Improve Medical Services Bolster Education Programs Improve Capacity for Public Health Programs Construction of Oyster Reefs Stipend for Business Adjustment to New Enviro Regulation Population and Demographic Studies Education and Training Funds for Increased Community Meetings/Programs Streamline Recovery Fund Procedure Language Translation Parish Administrative Capacity Herbivory Prevention
variables. Seven of nine environmental related variables met the criteria for an expanded
list. The community based expert group ranked herbivory prevention (0.0208) higher
than the other three expert groups and is within the 90% cumulative percent of model
variables. This ranking meets the criteria of for an optimal list.
The economic variables of outreach programs to promote tourism and fisheries
(0.33 8} and funding of government services (0.0241) met the criteria for an expanded list.
This expert group ranked four variables within the societal impact objective, bolster
education program (0.0 182), community outreach and support programs, population and
demographic studies (0.0075), and language translation (0.0029), in the lower third of
variables and below the inclusion criteria for an optimal list within this expert group.
Population and demographic studies and language translation were the two lowest
ranking variables in the prioritized list.
The variables within the logistical capacity for disaster response objective followed a
similar pattern in low prioritization as other expert groups. Education and training
(0.0337), basic equipment for residual oil removal (0.033), and streamline recovery fund
procedure (0.0197) ranked highest. The remaining variables in this objective group were
ranked below 90% of the cumulative percent for model variables. The trade-off scores
and cumulative percent of model variables for the community based expert group can be
found in Table 4-18.
96
Table 4-18:
Variable Cypress Swamp Restoration River Diversions Loans for Small Business Development Removal and Disposal of.Residual Oil Construction of Oyster Reefs Aquaculture Studies Improve Capacity for Public Health Programs Information Sharing Restoration of Barrier Islands Outreach Programs to Promote Tourism/Fisheries Education and Training Researching Impacts of Climate Change Basic Equipment for Residual Oil Clean Up Improve Medical Services Examine Sustainability of Fisheries Public Health Staffing and Surveillance Database Funding for Government Establish Monitoring System Increase Mental Health Capacity Herbivory Prevention Streamline Recovery Fund Procedure Bolster Education Programs Parish Administrative Capacity Stipend for Business Adjustment to New Enviro Regulation Community Outreach and Support Programs Monitoring Natural Attenuation of Residual Oil Funds for Increased Community Meetings/Programs Population and Demographic Studies Language Translation
prioritization for resource allocation. Though the value-trade off scores and cumulative
percent of the four expert groups will not be used in the created resource allocation tool,
they provide additional analysis to the differences in values of the expert group. In
97
summary, the economic stability sub-objectives were ranked high among all groups. To
provide more depth in assessing the prioritization of resilience measure per identified
objective, the following assessment utilizes the expanded list (80% of the cumulative
percent model). Four of the five variables were included within 80% cumulative percent
by three of four expert groups for each variable. Six ofthe environmental sub-objectives
were ranked within the 80% of the cumulative percent model by three of four expert
groups. Three variables from societal impacts followed the same ranking and just one
variable from the logistical capacity was ranked in the 80% of cumulative percent by
three of four expert groups.
4.6 Conclusion of Research Results:
The use of the Delphi method identified model variables for the resource allocation
tool. Fifteen experts participated in two rounds ofDelphi questionnaires of variable
selection and identified twenty-nine variables that met the inclusion criteria for model
variables. Thirty expert representatives of four identified expert groups (science and
technical, government representatives, business representatives, and community based
organizations) were approached to participate in ranking model variables. Participants
were asked to consider the use of funds from the BP deepwater horizon incident to
promote long-term resilience issues in their ranking of model variables.
The ranking procedure used in this research was the SMART model. Trade-off
scores were attained by averaging the study participant's response for each variable.
There were four objective variables and twenty-nine sub-objective variables. The
averaged response to each objective was multiplied by the averaged response to each
corresponding sub-objective within that objective. The attained trade-off scores for each
98
variable were sorted from the highest score to the lowest, thus creating the prioritized list.
The cumulative percent model was implemented, and cut-off points, established from
previous research, were implemented to identify a minimum, expanded, and optimal list
of resilience measures to be addressed in St. Bernard Parish with consideration of future
oil related incidents. The variables found in the minimum list (highest priority for
research objectives) are addressed below in table 4-19.
Table 4-19:
Minimum Variable List
Variables Trade-Off Score Cumulative Percent
Loans for Small Business Development Cypress Swamp Restoration Removal and Disposal of Residual Oil Restoration of Barrier Islands Aquaculture Studies River Diversions Outreach Programs to Promote Tourism/Fisheries Increase Mental Health Capacity Funding for Government Stipend for Business Adjustment to New Enviro Regulation Improve Capacity for Public Health Programs Construction of Oyster Reefs Improve Medical Services
environmental sub-objectives indicates a broad concern for the high rate of coastal
erosion and hurricane protection. Important to mention, the highest of prioritized
environmental variables associated with coastal erosion and hurricane protection do
create indirect resilience to future oil related incidents by stabilizing the ecosystem.
Kendal's Coefficient of Concordance (W) was used to assess the level of correlation
among all groups and within the expert groups. Levels of correlation were found to be
significant, but low among all study participants. The highest level of agreement was
found in ranking the societal impact variables with W = 0.342 and the lowest level of
agreement found in ranking the economic sub-objectives W = 0.142. A higher level of
agreement was found within the expert groups. The highest agreement came within the
business representative expert group. Levels of agreement ranged from W = 0.617 for
the economic stability variables to 0.354 for societal impact. The science and technical
expert group had the highest level of agreement (W = 0.434) for environmental variables
and the lowest level of agreement on economic stability variables at W = .282.
Community based groups had a level of agreement ranging from W = 0.494 (objective
ranking) toW= 0.235 for economic variables.
Overall the level of correlation on ranking of model variables was low. These results
can be interpreted in several fashions. The development of the survey and collection of
data for this research was completed as the full impacts of the oil related incident was
being assessed. Without complete and sound data on the extent ofthe impacts of the
Deepwater Horizon event, the experts are more likely to rank variables they feel most
relevant to the long-term resilience for St. Bernard. This may include measures that were
directly or indirectly impacted from the Deepwater Horizon event.
100
The null hypothesis of Kendall's Coefficient of Concordance was rejected for the
seven hypotheses in this research, indicating significance in the levels of correlation
(Tables 4-3 to 4-9). The Kruskal-Wallis H-test null hypothesis, indicating no difference
in median ranking among the previously mentioned study participant groups, was
accepted for the two hypotheses tested in this research (Tables 4-10 and 4-11 ).
Trade-off scores and cumulative percent for model variables were calculated for the
four expert groups. These results were not used in the creation of the final resource
allocation tool, rather to provide additional analysis and insight to the values and
preference of model variables within expert groups.
101
5.1 Overview of Results:
CHAPTERV
DISCUSSION
The views and values obtained through the SMART questionnaire are based on the
individual expert's personal experiences in dealing with the short comings of distribution
of funds and the needs of St. Bernard Parish following the BP Deepwater Horizon oil
leak of2010. Kendall's Coefficient of Concordance r,:w) was used to assess the
correlation in responses between the four expert groups as well as within each expert
group. The Kruskal-Wallis H-test was used to test the median ranking ofmodel variables
to assess if significant agreement existed between the experts. Additional statistical
analysis tests were conducted to assess correlation and agreement between the expert
group who identified the model variables during the Delphi method and participated in
the SMART questionnaire with those experts in expert groups who only participated in
the SMART questionnaire. The following analysis answers research question number
five: What can be learned through the development of this model?
The issues contributing to the lack of resilience in St. Bernard Parish are complex and
significant. Major issues facing the Parish are slow recovery due to weakened resilience
from past disasters, the net loss of sediment leading to coastal erosion (as a result of
extensive levee systems in Louisiana), subsidence, salt water intrusion, decreased storm
protection as a result of an unhealthy deltaic system, a comparatively economically
depressed region in the United States, and lack of adequate public health infrastructure.
102
Though this study asked expert participants to identify measures that will build
resilience against future oil related incidents in the Gulf of Mexico through utilizing the
Delphi method, many identified variables were not directly associated to oil related
incidents. The variables not pertaining to future oil related incidents, however, will
contribute towards creating an increased general resilience by improving the diversity of
economic opportunities, creating and rehabilitating a fragile ecosystem, and improving
quality of life for the residents of St. Bernard Parish. This is likely a result that occurred
due to the expert participants concerns about the long-term viability of the Parish and a
realization that the threat of future oil related incidents in the Gulf is one of many
resilience issues needing to be addressed in the Parish. In an area that is vulnerable to
both man-made and natural disasters, these measures will provide means for creating
resilience, and mitigating the initial and long-term impacts associated with future
disasters. Those identified variables that establish general resilience measures in St.
Bernard Parish addressed issues such as coastal erosion and strengthening natural and
man-made resources against future disasters.
Surprisingly, a major issue that impacts the sustainability and economic viability the
Louisiana Gulf Coast was not addressed. The hypoxic zone found off the Louisiana coast
is currently the size ofNew Jersey and Delaware. This hypoxic zone (a zone that has less
than 2mg/L of dissolved oxygen), is a result of excess nutrient flow associated with
Midwestern agriculture run-off in the Mississippi River, and creates an environment that
is inhospitable to nearly all aquatic species. In terms of building long-term resilience to
eco-system services in the Gulf of Mexico, measures addressing and mitigating the
hypoxic zone (dead-zone) must be established and implemented.
103
As the Mississippi Delta continues to degrade, St. Bernard becomes increasingly
dependent on oil and gas operations and fisheries as means for economic stability. Both
sectors have a high degree of dependence on the environment for their operations and are
susceptible to major slow down should a disturbance cause shock to the environment.
This dependence on such few opportunities for economic livelihood and the dependence
of which those opportunities have on the declining health of the ecosystem is a core issue
highlighting the lack of resilience in St. Bernard Parish. The recognition of the
interconnectedness between the economic viability of St. Bernard and a healthy
ecosystem was indicated by expert responses through their value trade-offs. The highest
rated variables in the created prioritized list were those associated with the environmental
and economic stability object groupings.
The elicited responses from expert representatives of expert groups ranked the
environmental objective most important, accounting for 39.04% of total importance
followed by the economic stability objective (27.44%). The societal impacts objective
was ranked third in importance at 20.31%, and the logistical capacity for disaster
response was ranked as the least important of the four objectives (13.22%).
Interestingly, in addition to selecting resilience measures that promote general
resilience by the fifteen experts who identified model variables (Delphi method), those
measures were highly prioritized by the additional twenty-one experts who participated in
the SMART portion ofthe study. This further illustrates the view of the expert
participants in the importance to approach resilience in St. Bernard in a generalized
manner. Resilience is a concept that spans many facets of society and building resilience
in a more general sense will provide an ability for the community to recover quicker from
104
any disaster, in contrast to a specific disaster, such as future oil related incidents in the
Gulf. Measures such as cypress swamp restoration, construction of berms, and loans for
small business development will create a more generalized approach towards building
resilience to future man-made or natural disasters in St. Bernard Parish.
Resilience measures associated with logistical capacity for disaster response
following an oil related incident had the lowest prioritization among the four sub
objective groups. This further underscores the sentiment among expert participants to
allocate available funds towards general resilience measures that will prepare St. Bernard
Parish for future man-made or natural disasters opposed to specifically addressing
resilience towards future oil related incidents.
Model variables that were ranked highly were primarily dealing with the continued
fallout and on-going concerns with the impacts associated with the BP Deepwater
Horizon incident. This is illustrated in the high rankings of removal and disposal of
residual oil, aquaculture studies, outreach programs to promote tourism/fisheries, and
stipends for business adjustment to new environmental regulations. This view by expert
participants indicates that priorities addressing impacts associated with the Deepwater
Horizon event and focusing on generalized resilience in the Parish (as opposed to
specifically preparing for a future oil related incident) as the most beneficial approach
towards addressing the long-term viability of St. Bernard Parish.
An alternative explanation to the identification and high prioritization of resilience
·measures not directly associated to future oil related incidents in the Gulf of Mexico is
the timing in which the questionnaires and data were collected. The data collection phase
105
of this research was conducted soon following the BP Deepwater Horizon incident. The
size and nature of this type of incident had never occurred in the Gulf of Mexico and the
long-term impacts associated with the incident are still unknown. The timing of this
project may have resulted in the selection of variables not directly associated with the BP
Deepwater Horizon incident and provides an explanation of why some identified
variables did not have a direct relation to building resilience in St. Bernard Parish.
5.2 Hypothesis Testing:
The hypothesis testing revealed interesting insight on the values and importance
placed on model variables. This section will provide an analysis of Kendall's Coefficient
of Concordance (W) for all experts and the expert groups. Low levels of agreement on
the prioritization of resource allocation for model variables was found among study
participants. A potential explanation for this result is the broad and diverse nature of this
study and study participants that represent a wide range of professions. The participants
had varying, and at times conflicting, views on what was of importance to the long-term
sustainability of St. Bernard Parish. The conflicting rankings by participants indicate a
possible lack of understanding and or recognition of the integrated role the coastal
ecosystem plays in the economic stability of St. Bernard Parish. The low level of
agreement among all study participants indicate a disconnect between expert groups in
identifying and establishing a cohesive approach towards long-term resilience in St.
Bernard Parish. This is of particular concern due to the critical and immediate resilience
measures needed to maintain viability in St. Bernard. Cohesive political will is required
to promptly address and implement resilience measures
106
The highest level of agreement was found in the ranking of societal impact sub
objectives at W=0.325. This result was not anticipated; however it indicates a cohesive
view among the expert groups in approaching the societal based needs of St. Bernard
Parish. Although the societal impacts sub-objectives were the most agreed on by study
participants, the societal impact objective was prioritized behind the environmental and
economic stability objectives. The economic sub-objectives had the lowest level of
agreement at W= 0.142, yet was ranked high in the resource allocation list, of which one
was the highest ranked variable: loans for small business development.
Analysis of agreement within each of the four expert groups indicated a higher level
of agreement within each group when compared to levels of agreement when all thirty
participants were placed in one group. This result indicates study participants within an
expert group with similar backgrounds and training tend to have the same perception in
prioritization of variables and the needs of St. Bernard Parish.
5.3 Review of Objective Ranking:
Significance in association of ranking of model objectives was not found for the
government expert group, the impacted business expert group, and the science and
technical expert group. Significance levels were established at a p-value of0.05.
There are several reasons as to why there was lack of significance. The first is that at
the time of data collection, the full impacts of the BP Deepwater Horizon were unknown.
This likely led experts to select resilience measures they felt important towards building
resilience for storm protection and coastal erosion. These measures build general
resilience in St. Bernard Parish against future man-made or natural disaster scenarios, and
107
therefore indirectly build resilience for future oil related incidents. The second reason is
likelihood that experts' personal political viewpoints contributed to lack of significance
associated with correlation among experts in ranking objectives in this research. It is
observed that environmental issues have become highly politicized in this country.
The low levels of correlation (Kendall's Coefficient of Concordance) among ranking
the environmental variables as well the lack of agreement found in the median rank of
environmental variables (Kruskal-Wallis H-test), indicate the complexity and controversy
behind implementation of the environmental restoration variables. As environmental
related issues have become politicized in the United States, there is a likelihood that
experts political viewpoints contributed to the lower levels of correlation in Kendall's
Coefficient of Concordance in objective ranking and disagreement in the Kruskal-Wallis
H-test agreement in median rank of model variables. This is extrapolated as the level of
agreement found within the business representative group had high levels of agreement in
pro-economic growth variables, yet differed in priority from the remaining three expert
groups. In the current political climate, pro-economic growth ideology tends to be closer
associated with conservative party recommendations. This study did not confirm the
political affiliation of each expert participant.
An issue impacting p-values are the sample sizes of the individual groups. The
groups that lacked significance had a smaller sample sizes (n = six for the Government
Representative expert group, six for the Business Representative expert group, and seven
for the Science and Technical expert group). It is likely that with larger sample sizes, the
p-values would be lower, thus leading to significance in the statistical analysis.
108
5.4 Review of Environment Objective and Sub-objectives and Results:
The environment was ranked as the most important objective in the decision model
with a ranking of trade-off score of 0.3904 for all thirty study participants. Six of nine
sub-objectives related to this objective are part ofthe optimal list of the cumulative
percent of75%. The highest ranked variable was cypress swamp restoration, followed by
removal and disposal of any residual oil that is found, then restoration of barrier islands,
followed by river diversions, construction of oyster reefs, and examination of the
sustainability of the fisheries. At the outset of this study, the researcher's hypothesis was
that the environmental objective would be considered by study participants to be the most
important one to address and its corresponding sub-objectives would be highly ranked.
The interview sessions and questionnaire responses indicated a strong understanding and
sentiment to the importance of environmental protection from both man-made and natural
disasters and restoration in St. Bernard Parish to maintain its long-term viability.
Significance in association in ranking environmental variables was established by all
groups. The levels of association identify a fundamental difference in which resilience
measure should be implemented first. The nature and economic utility of coastal
Louisiana ecosystem is complex. Experts with an economic stake in the development
and health of certain aspects of the ecosystem, which benefit their livelihood, are more
likely to place a higher prioritization on variables that promote their specific
field/industry. For example, those with a stake in the oyster industry had oyster reef
construction and river diversions highly prioritized.
109
5.5 Review of Logistical Capacity Objective and Sub-Objectives and Results:
The variables in this objective were selected to gauge the allocation of funds towards
establishing a preparedness plan and having all the necessary resources available for the
smooth response to a large-scale oil related event. The logistical capacity for disaster
response was the lowest rated objective accounting for 13.22% of the overall importance
of model objectives. The sub-objectives as a group were subsequently ranked low as
well. The highest ranked variable, and the only variable that fell within the top 80%
limit of the cumulative percent model to be included in the expanded list was the sub-
objective for information sharing (3.04%) followed by basic equipment for residual oil
clean up, establishing monitoring systems, education and training, parish administrative
capacity, and funds for increased community meetings and programs. The lowest ranked
variable in this objective was ranked second lowest on the complete list of resource
allocation. These results indicate a confidence of study participants in the development
of a preparedness plan on a state and local level and the handling of the execution phase
of the plan.
5.6 Review of Economic Objective and Sub-Objectives and Results:
The economic stability objective group was prioritized second behind the
environmental objective accounting for 27.44% ofthe overall importance. This objective
had all five sub-objectives ranked within the top 50% of the cumulative percent model,
including the top overall variable of loans for small business development. On the outset
of the study the researcher anticipated the sub-objectives within the economic stability
objective to be ranked high, as mentioned in chapter four. The results capture the general
110
sentiment of the region's priority on economic development and job growth. It was
mentioned by several study participants the importance of innovation throughout coastal
Louisiana as a way to maintain the long-term viability ofthe region. Further, the
comments suggested that as the opportunities of employment are primarily limited to oil
and gas, tourism, and fisheries along the gulf coast, a strong effort was needed to
encourage citizens to create jobs through innovation. The participants who commented
on this did not provide further details on which sector they would like to see have an
increased presence in the region. The overall level of correlation by all thirty participants
was very low (0.142). This can potentially be attributed to the diverse focus of expert's
professions and perceptions on economic measures that will optimize growth. The
Business representative expert group had the highest level of correlation in this set of
sub-objective. This is likely a result that most participants in this expert group were
business executives whose experience in focusing on predicting needs and changes in the
local business atmosphere in order to maintain a profitable firm.
5. 7 Review of Societal Impacts Objective and Sub-Objectives and Results:
The societal impacts resulting from an oil related event similar to that of the
Deepwater Horizon oil leak on St. Bernard Parish is significant. This objective was
ranked with the third highest importance behind the environmental objective and
economic stability objective at 0.2031. Four of the seven variables ranked within the
minimum list, with the highest ranked variable being a focus on mental health. Residents
of St. Bernard are still recovering from the psychological trauma associated with
Hurricane Katrina (GAO, 2009, Rhodes, et. al. 2010). Though results from mental health
studies following the Deepwater Horizon incident have not been reported, mental health
111
studies following the Exxon Valdez spill in Prince William Sound have indicated a
likelihood of mental health impacts from the BP incident (Gill, 2011 ). Over twenty years
later, mental health issues are still pervasive as a result of the Exxon Valdez oil spill. St.
Bernard Parish set a priority to upgrade mental health and medical facilities in the Parish
to adequately handle the need following Hurricane Katrina as well as for any future
events (Mitchell, 2007). Unfortunately, mental health facilities are limited and there is no
hospital in St. Bernard. The remaining variables were to improve capacity for public
health programs, improve medical services, and community outreach and support
programs.
5.8 Policy Implications:
The BP Deepwater Horizon incident was the first major oil related incident occurring
in a body of water in the United States since the Exxon Valdez oil spill in 1989 which
resulted in the release of over 250,000 barrels. The recent BP incident highlighted flaws
in the oil pollution act of 1990 (OPA), which was legislation created as a result of the
Exxon Valdez; it is designed to address oil tanker spills and not off-shore drilling events.
Thirty percent of the United States national oil production comes from drilling operations
in the Gulf of Mexico (DOE, 201 0) , and there is a likelihood of a similar incident
occurring in the future. Policies need to be updated to address and to include current
national status of oil and gas production. The response to oil spill legislation should
approach different ecosystems, spill or leak sizes, and take an integrated approach to
restoration in affected areas.
112
The OPA fmes are levied by taxing the responsible party 18.75% per barrel at the
average barrel market cost. OP A fmes resulting from the Deepwater Horizon oil leak
would cost BP approximately 78.1 million US dollars (USD). This number is based on
4.9 million barrels of oil leaked, with a market average of 85 USD per barrel taxed at
18.75%. The price per barrel is likely to change based on the fluctuating market price of
oil.
In addition to the OPA fmes, the Clean Water Act fines would be assessed as well.
Current legislation dictates the CW A fme to be levied on the responsible party per barrel
of oil leaked. The fine per barrel ranges from $1,100 to $4,300 (Federal CWA). The
specific fine relates to the court's ruling on levels of gross negligence. The resulting
fmes from court ruling are currently distributed to the National Oil Spill Trust Fund and
the United States Treasury.
Following the prospectus defense for this research, The Resource and Ecosystem
Sustainability, Tourist Opportunities, and Revived Economy of the Gulf Coast Act of
2011 (RESTORE the Gulf Act of2011) was introduced by a bi-partisan group of
Senators from the Gulf States (Mary Landreau, Thad Cochran, Jeff Sessions, and Kay
Hutchinson). If passed, the legislation would require that 4/5 (80%) of funds from the
Clean Water Act funds be distributed to the five Gulf States that were affected by the BP
Deepwater Horizon spill. This legislation would create a subsection in the Federal Water
Pollution Control Act to allow for the allocation of CW A funds to the Gulf Coast. The
subsection indicates that 35% of the funds are to be equally distributed to the five Gulf
States .. Sixty percent (60%) ofthe funds would be allocated the Gulf Coast Ecosystem
Restoration Council, and 5% of funds would be distributed for research and science
113
--- ----------------------------
programs in the Gulf coast region. Of the 35% that is to be equally distributed to each
state, it is up to the Governor of each state as well as state agencies to decide on the
appropriate distribution ofthe funds (Environmental Defense fund, 2011). Currently
there is no method described by the National Oil Spill Commission or the Gulf Coast
Ecosystem Restoration Task Force to prioritize tasks that need to be addressed with the
recommended distribution of funds (Final report, 2011 ).
The resource allocation tool developed in this research approaches this complex
problem through a recognized method in the decision sciences. The created tool is only
relevant to the needs of St. Bernard Parish, LA., where the research project took place.
Expert groups in different Gulf of Mexico coastal areas may respond and provide
different results for their particular geographical area.
The decision analytic tool developed by this research should be considered by policy
makers in Baton Rouge, LA and state government agencies as a recommendation to the
distribution of funds. As dictated by the RESTORE the Gulf act, The Gulf Coast
Ecosystem Restoration Council will receive 60% of the CWA funds. A principle
objective for the council is to develop a comprehensive plan addressing restoration and
protect ecosystems, natural resources, wetland.s, and natural resources (RESTORE the.
Gulf, 2011). In October, 2011, the Gulf Coast Ecosystem Restoration Task Force
released its preliminary report, Gulf of Mexico Regional Ecosystem Restoration Strategy
on the issues facing the Gulf Coast. The report is anticipated to be used for guidance to
the Council should the RESTORE act be made into law. The resource allocation tool can
be utilized by the council to address restoration issues as well as the preparedness of St.
Bernard Parish in the event of a future oil related event in the Gulf of Mexico.
114
The second use of this research in the RESTORE the Gulf Act of2011 is the
distribution of 35% of CW A funds to the five Gulf States impacted by the Deepwater
Horizon event. The allocations of funds will be distributed equally to the five Gulf
States. The use of this fmancial resource is to be allocated at the discretion of the state
government (the Governor's office and state agencies). The likely distribution of funds
for Louisiana will be made in close consultation with the Office of Coastal Protection and
Restoration and referencing the State Master Plan. The Louisiana State Master Plan was
developed to address the critical needs of coastal Louisiana in 2007. It is updated every
five years to address the quickly changing needs and issues facing coastal Louisiana. The
resource allocation tool can be considered as a method for prioritizing the allocation of
limited monetary resources in the 2012 State Master Plan. In the event that the
RESTORE the Gulf Act of2011 (or future legislation addressing the Clean Water Act
fines to be allocated towards resilience efforts in the Gulf) does not pass into law, the
resource allocation tool remains relevant in consideration for future updates of the
Louisiana State Master Plan.
The State Master plan of Louisiana is scheduled to be released for public comment in
March of2012. The Master Plan is designed to approach the coastal needs of Louisiana
and is updated every five years. Though this plan has indirect economic benefits to
Louisiana citizens, The State Master Plan focuses on environmental projects to reduce
coastal erosion, promote the wetlands, and an overall healthy ecosystem. The anticipated
funding for the 2012 State Master Plan comes from $80 million per year from Coastal
Wetlands Planning, Protection and Restoration Act (CWPPRA), $110 million per year
from the Gulf of Mexico Energy Security Act (GOMESA), and $150 million per year
115
from Louisiana Coastal Area (LCA) Program (Draft State Master Plan F AQ, 2012). The
finances associated with implementation of projects in the State Master Plan are
anticipated, but not guaranteed.
5.9 Public Health Pertinence:
Oil and gas exploration and production provide an important source of economic
stability in Louisiana for many years. The role of oil and gas operations will play an
increased importance to Louisiana as sustained political pressure in the United States
works towards energy independence. With oil related operations above 3,800 in 2011 in
the Gulf of Mexico and anticipated to increase, the probability of a future oil related
incident is high.
A cornerstone of the public health profession is the emphasis on preparedness and
prevention. Allocating funds from the BP oil leak towards long-term resilience issues in
St. Bernard Parish will strengthen the environment and coastal communities'
preparedness to mitigate damages associated with such a future event. For example, had
. the identified model variables in this research been implemented prior to the Deepwater
Horizon incident, impacts of the moratorium on oil related operations and fisheries would
have been reduced due to the diversification of employment opportunities, less oil would
have reached the coast due to the community being better suited to handle an oil related,
and the environmental impacts on the coastline would have been mitigated. Addressing
the lack of resilience and preparedness in coastal Louisiana and other. gulf states is a
concept addressed in the National Commission on the BP Deepwater Horizon Oil Spill
and Offshore Drilling final report and the preliminary report by the Gulf Coast
116
Ecosystem Restoration Task Force. A study on the validity and cost-effectiveness of
allocating resources to preventative measure in community based projects was addressed
by the Multi Mitigation Council and concluded that for every one dollar spent towards
hazard mitigation equates to a four dollar saving (MMC, 2005). In order to optimize
resilience and preparedness in coastal communities, an effective strategy to prioritize the
distribution of monetary funds must be implemented
The decision analytic tool created in this research provides a prioritized list of
resilience measures that promote resilience to future oil related incidents as well as
establishes generalized resilience in St. Bernard Parish. Monetary resources are limited in
these situations and it is critical to efficiently utilize the limited funds. Prior to
implementation of any measures associated with this research, a cost assessment must be
conducted for each variable. When the cost of implementation is accounted for, the
policy maker must decide on which measures are feasible to implement with the available
funds. For example, if there are three resilience measures with lower priority that can be
implemented at a cost equal to that of a higher priority, the decision maker must decide
on which choice will create greater resilience. Further, an assessment of projects found
on this prioritized list with secured funding must be identified to minimize duplication of
projects.
The benefits of implementing the prioritized list of resilience measures created in this
research in the short term are large. Job creation through the small business loans as well
as the implementation of the construction and labor employment resulting from the
implementation of many variables will provide an economic boost to the region. The
correlation between the economic stability of a community and the health of the
117
community is strong (Bloom, 2008). Upon the completion of implementing the selected
measures, the increased resilience will mitigate the impacts of any future disasters in the
Parish. This will reduce the physical and psychological, as well as ecological and
economic impacts a future disaster has in St. Bernard Parish.
Many of the variables in the optimal list promote coastal restoration. The restoration
projects will rehabilitate a depleted ecosystem, leading to greater bio-diver~ity and a
reduction of environmental factors causing degradation such as salt-water intrusion and
coastal erosion. The resulting measures will benefit St. Bernard ecosystem services by
creating a healthier ecosystem and better subsequent productivity in the commercial and
tourist fisheries sector. Further, these measures will build resilience to the impacts of
future anthropogenic or natural disasters. The implementation of the societal needs will
provide basic services, mostly pertaining to health and education that are needed in St.
Bernard.
5.10 Conclusion and Summary:
Coastal Louisiana's vulnerabilities to natural and man-made disasters continue to be
exposed. Disasters, such as the BP Deepwater Horizon incident, further degrade the
environment; weaken the strength and resources of communities, and impact individuals'
health and well-being. With the recent passing of the RESTORE the GULF Act of2011,
significant quantities of financial resources (5.3 to 21 billion dollars pending on level of
negligence identified by the Federal Court in New Orleans, LA) resulting from civil
penalties associated the Clean Water Act will be allocated towards building resilience in
Gulf of Mexico states.
118
The results of the research conducted in this study should be considered by policy
makers at the state and federal level as an effective method to identify and prioritize
resilience measures for St. Bernard Parish. However, it is essential that prior to the
implementation of resilience measures to conduct a cost assessment of each measure.
The recommended and highest prioritized resilience measures in this research will likely
come with significant costs. Upon completion of the cost assessment, the policy maker
must consider which resilience measure is feasible to implement and/or if it would be a
better utilization of monetary resources to fund multiple projects that do not have the
highest of prioritization but deemed more cost effective~ Further, prior to implementation
the policy maker must assess if funding has been secured to implement any resilience
measures that have been identified with in the prioritized list found in this research.
The research created the prioritized list by sorting resilience measures with the
highest trade-off score (representing highest priority) to the lowest trade-off score
(representing the lowest priority). Based on previous research that utilize the SMART
methodology (Mack, 2008, Jirapongsuwan, 2008, Zornes, 2007), the cumulative percent
model was implemented to establish three lists: Minimal list, Expanded list, and Optimal
list. The aforementioned list provides the policy maker with a reference point in which to
approach implementing tasks based on the availability of funds. Table 5-l illustrates the
breakpoint for each list.
119
Table 5-1:
List Cumulative Percent
Minimum 0-70%
Expanded 0-80%
Optimal 0-90%
The available funds resulting from BP penalties to be utilized for resilience measures
in St. Bernard Parish will not be adequate to address all identified resilience measures.
Pending the outcome of litigation, policy makers are recommended to utilize the
appropriate list (minimum, expanded, or optimal) to implement according to available
funds. The breakdown of the lists are as follows:
Minimum Variable List
Variables Trade-Off Score Cumulative Percent
Loans for Small Business Development Cypress Swamp Restoration Removal and Disposal of Residual Oil Restoration of Barrier Islands Aquaculture Studies River Diversions Outreach Programs to Promote Tourism/Fisheries Increase Mental Health Capacity Funding for Government Stipend for Business Adjustment to New Enviro Regulation Improve Capacity for Public Health Programs Construction of Oyster Reefs Improve Medical Services
Community Outreach and Support Programs Examine Sustainability of Fisheries Information Sharing
Optimal Variable List
0.0336 0.0331 0.0304
71.97 75.28 78.32
Variables Trade-Off Score Cumulative Percent
Public Health Staffmg and Surveillance Database Monitoring Natural Attenuation of Residual Oil Basic Equipment for Residual Oil Clean Up Establish Monitoring System Bolster Education Programs
Low Ranking Variable List
0.0244 0.0239 0.0237 0.0215 0.0204
80.76 83.15 85.52 87.67 89.71
Variables Trade-Off Score Cumulative Percent
Streamline Recovery Fund Procedure Researching Impacts of Climate Change Education and Training Herbivory Prevention Parish Administrative Capacity Population and Demographic Studies Funds for Increased Community Meetings/ Programs Language Translation
TO: Benjamin Schulte, r-.lP.H. FROt-.-1: Tulane University Sociai-BehaviomiiRB
STUDY TITLE: [2:35200-'1] Deepwater Horizon Oil Leak: A Decision Analytic Approach to Resource Allocation
IRS REFERENCE"'!-: 11-22.5200U
SUBMISSION TYPE: New Project
ACTION: APPROVED
IRS APPROVAL DATE: September 20. 2011
IRS EXPIRATION DATE: September 19. 2012
Thank you for your recent New Project submissic•n. The Tulane University Institutional Review Board has granted approval for the above-referenced protocol together with:
Amendment!Modification Form (UPDATED: OB/19!20'11)
Application for Hum<:ln Subjects Resemch, Part 'I (UPDATED: 06;22/20'1'1)
Application Part 2 (UPDATED: 09i12i201'!)
Consent Fom1, Version Date f•il2f1l (UPDATED: 09.'12!20'1 ·1)
in accordance with 45 CFR 413.'1'l 0 . Please note the expiration date of the protocol <Jbove.
125
Appendix B: Delphi Study Round One
Please answer these questions based upon your expertise gained following the BP Deepwater Horizon oil disaster and understanding of regional needs for St. Bernard Parish:
1. Please rate each item on a scale of 1-5 according to its importance in maximizing environmental, economic, and societal recovery in St. Bernard Parish. Circle (or bold) the best answer. 5 = strongly agree, 4 = agree, 3 = neither agree nor disagree, 2 = disagree, and 1 = strongly disagree.
2. Please list any additional important variables that should be taken into account that promotes environment, economic, and societal recovery for St. Bernard Parish following the BP Deepwater Horizon and rate those statements on the aforementioned scale.
Definition of terms
Resilience: The capacity of a system to absorb disturbance and reorganize while undergoing change so as to still retain essentially the same function, structure, identity, and feedback modify
Regional economic stability: Methods and practices that promote the economic development and growth of a region. In this research it pertains to St. Bernard Parish.
Environmental restoration: Practices that promote ecological and biological recovery following a man-made or natural disaster.
Societal measures: Efforts to promote the wellbeing and health of a community affected by a man-made or natural disaster.
126
Environment:
Removal and disposal of residual oil: Allocate resources to cleaning any oil that should surface, wash-up to shore, or invest in technology that best addresses underwater oil plumes.
Construct Oyster Reefs: Oyster reefs would promote shoreline stability and contribute to the ecosystem of Lake Borgne.
Cypress Swamp Restoration: Allocate funds to reroute the use of wastewater and freshwater to flow into the cypress swamps to mitigate the impacts of saltwater intrusion and promote cypress swamp restoration.
Prevention of Herbivory: Trapping of invasive species, primarily nutria, to reduce their impact on wetlands, therefore providing a healthier and more resilient wetland system.
Restoration of Barrier Islands: Infusion of sediment to counter the landmass lost to the barrier islands associated with erosion.
River Diversions: Use available funds for river diversions (Canarvon and Violet) in St. Bernard Parish to promote the introduction of freshwater to the marshes in St. Bernard Parish.
Sustain ability of fisheries: Conduct studies quantifying the impact of the oil leak on fish populations and analyzing the long-term impact.
127
Logistical Capacity for Long Term Recovery:
Basic Equipment and Necessities for residual oil clean-up: Boom, boats, dispersants, Personal Protective Equipment, food and shelter for workers, fuel.
Education and Training: HAZWOPER training, and other related training for programs to those forced to adjust their occupation to relevant and available work following an oil leak.
Establish monitoring system: Environmental Sensitivity Index, use of GPS, and methods that allow ease of reporting of oil sightings.
Information Sharing: Develop and implement systems for data integration, synthesis, sharing and dissemination.
Streamline Recovery Fund Procedure: Provide additional resources to community, particularly workers who can verbally describe the paperwork that is required with in applications.
Parish Staff I Parish Administrative capacity: Provide funds for additional St. Bernard Parish staff and administration to handle increase administrative needs associated with the BP Deepwater Horizon incident.
Fundingfor increased community meetings and programs: Promote understanding of all aspects of recovery and restoration of a disaster between stakeholder groups.
128
Regional Economic Stability:
Outreach Programs: Create television commercials and magazine adds to promote tourism.
Aquaculture Studies: Finance for independent laboratories to examine extent of contamination and bioabsorptionlbioaccumulation of aquaculture.
Restaurant and Hotel Subsidies: Subsidies for restaurant and hotel owners and workers to account for 100% of 5-year average profit prior to disaster.
Stipend for business adjustment to new environmental regulation: Provide money to small business that funds updating their business to meet new environmental regulation resulting from oil spill.
Loans for small business development: Increase availability of micro loans to individuals for small business use and/or start-up from affected communities.
Government funding: Provide local governments with funds to compensate reduced tax revenue as a result of reduced industry output.
Societal Impacts:
Community education outreach: Provide outreach sessions to communities that provide scientific backing on the health outcomes and effects of the oil and dispersants.
Bolster Educational Programs: Promote education and literacy in rural and urban regions.
129
--~--~-- --------~-------~---------~
Public health staffing and databases: Increase states resources for surveillance of adverse health effects associated with oil and dispersants.
Improve capacity of public health programs: Increase funds for social programs such as food stamps, school lunches, immunizations, and maternal health for effected populations.
Increase mental health capacity: Develop resources and providers to deal with mental health issues, like Post Traumatic Stress Disorder (PTSD), to communities affected by the spill. Please see Appendix A for the decision tree.
Population and demographic studies: Establish baseline data following the oil disaster to help determine future resource allocation.
Language translation: Provide language translation services to those in need.
Please answer these questions based upon your expertise gained following the BP Deepwater Horizon oil disaster and understanding of regional needs for St. Bernard Parish:
3. Please reconsider your own previous rating and that of the entire panel (the Round 1 group mean and standard deviation).
4. Rate each item on a scale of 1-5 according to its importance in maximizing environmental, economic, and societal recovery in St. Bernard Parish. Circle (or bold) the best answer. 5 = strongly agree, 4 = agree, 3 = neither agree nor disagree, 2 = disagree, and 1 = strongly disagree.
5. Please list any additional important variables that should be taken into account that promotes environment, economic, and societal recovery for St. Bernard Parish following the BP Deepwater Horizon and rate those statements on the aforementioned scale.
Definition of terms
Environmental Variables: Measures that promote restoration and ecological recovery following an oil related incident.
Economic Variables: Methods and practices that promote economic development and growth in St. Bernard Parish following an oil related incident.
Logistical Capacity for Disaster Response Variables: Measures that promote the operational capacity to handle future oil related incidents.
Societal Impacts Variables: Measures to promote the well-being and health of communities following an oil related incident in St. Bernard Parish
Mean: The mathematical average of all scores submitted for each variable
Standard Deviation (Std Deviation): A statistic that shows the spread or dispersion of scores for a particular variable. The more widely the scores are spread out, the larger the standard deviation. The lower the score, the stronger the agreement.
139
Example: A standard Deviation of 0.5 reflects a stronger agreement in response than a score of 1.0
Please complete and return round 2 questionnaire by September 13th, 2011
If you have any questions, or concerns, please feel free to contact me at 312.402.4704 or Thankyou
Environment:
Removal and disposal of residual oil: Allocate resources to cleaning oil that should surface, wash-up to shore, or invest in technology that best addresses underwater oil plumes.
Construct Oyster Reefs: Oyster reefs would promote shoreline stability and contribute to the ecosystem of Lake Borgne, St. Bernard Parish.
Cypress Swamp Restoration: Allocate funds to reroute the use of wastewater and freshwater to flow into the cypress swamps to mitigate the impacts of saltwater intrusion and promote cypress swamp restoration.
Prevention of Herbivory: Trapping of invasive species, primarily nutria, to reduce their impact on wetlands, therefore providing a healthier and more resilient wetland system.
Restoration of Barrier Islands: Infusion of sediment to counter the landmass lost to the barrier islands associated with erosion.
River Diversions: Use available funds for river diversions (Canarvon and Violet) in St.
Bernard Parish to promote the introduction of freshwater to the marshes in St. Bernard Parish.
140
Sustain ability of fzsheries: Conduct studies quantifying the impact of the oil leak on fish populations and analyzing the long-term impact.
Logistical Capacity for Long Term Recovery:
Basic Equipment and Necessities for residual oil clean-up: Boom, boats, dispersants, Personal Protective Equipment, food and shelter for workers, fuel.
Environmental Health Education and Training: HAZWOPER training, and other related training for programs to those needing to adjust their occupation to relevant and available work following an oil leak.
Establish monitoring system: Environmental Sensitivity Index, use of GPS, and methods that allow ease of reporting of oil sightings.
Information Sharing: Develop and implement systems for data integration, synthesis, sharing and dissemination.
Streamline Recovery Fund Procedure: Provide additional resources to community, particularly workers who can describe the paperwork that is required with in applications.
Parish Staff I Parish Administrative capacity: Provide funds for additional St. Bernard Parish staff and administration to handle increase administrative needs associated with the BP Deepwater Horizon incident.
Funding for increased community meetings and programs: Promote understanding of all aspects of recovery and restoration of a disaster between expert groups.
141
Regional Economic Stability:
Economic Outreach Programs: Create television commercials and magazine adds to promote tourism.
Aquaculture Studies: Finance for independent laboratories to examine extent of contamination and bioabsorptionlbioaccumulation of aquaculture.
Restaurant and Hotel Subsidies: Subsidies for restaurant and hotel owners and workers to account for 100% of 5-year average profit prior to disaster.
Stipend for business adjustment to new environmental regulation: Provide money to small business that fund updating their business to meet new environmental regulation resulting from oil spill.
Loans for small business development: Increase availability of micro loans to individuals for small business use and/or start-up from affected communities.
Government funding: Provide local governments with funds to compensate reduced tax revenue as a result of reduced industry output.
Societal Impacts:
Community education outreach: Provide outreach sessions to communities that provide scientific backing on the health outcomes and effects of the oil and dispersants.
Bolster Educational Programs: Promote education and literacy in rural and urban regions.
Public health staffing and databases: Increase states resources for surveillance of adverse health effects associated with oil and dispersants.
142
Improve capacity of public health programs: Increase funds for social programs such as food stamps, school lunches, immunizations, and maternal health for effected populations.
Increase mental health care capacity: Develop resources and providers to deal with mental health problems, like Post Traumatic Stress Disorder (PTSD), to communities affected by the spill.
Population and demographic studies: Establish baseline data following the oil disaster to help determine future resource allocation.
Language translation: Provide language translation services to those in need.
143
Strongly Agree 7 Strongly Std
Variables Disagree Mean Deviation
Promote Regional Economic Stability
Outreach Programs 5 4 3 2 1 4.00 1.00
Aquaculture Studies 5 4 3 2 1 3.80 0.941
Restaurant and Hotel Subsidies 5 4 3 2 1 2.93 0.798
Stipend for Business Adjustment to New Environmental Regulation 5 4 3 2 1 3.86 0.915
Loans for Small Business Development 5 4 3 2 1 4.33 0.899
Funding for Government Services 5 4 3 2 1 3.46 1.06
lmJlrQVe Me>:!kal Sen1CC:! I 2 3(£) 0..2416 8 lo 0,01.!!5 2 (£) 0.2857
.R,\WD,\TA
SCIENCE AND TECHNICAL REPRESENTATIVES EXPERT GROUP I uJl~rt 13 1 E>J•"'rl 14 1 t:.qo~rt 15
Objecth"C• i Rank Raw We\'!b!l Rank RAW WdshC I ll.W: R~w Wei.!;hl I f:nviromnC!It!l I I 4() 04210 1 J 12 0.1875 1 I 300 o 769~ 1
Loginical C~i1)• for Diwa Respomc I 4 10 o 1052 1 I 27 04218 1 3 20 o os12 1 ltesioml Ecoocwnk Sut.Dit~ I 2 3(1 OJIS'I 1 2 15 01:>43 1 2 60 O.LHS I SociotillmpaGH I 3 15 0.1578 1 4 10 0.1.562 1 " 10 0.02.56 I
I RemQV&I AJid Di•JIO~I QfRe~xl~ Oil I I 600 035!/J 1 2 IO&J 02070 1 I 11520 o_~"N 1
C)'prCSII Swamp Rcalor.Wm I 6 45 (1.0266 1 7 45 0.0086 I 4 %0 0.04:"9 I Coo:llndion of()yg!ef Rcefl I 9 JO 0.0059 1 6 ~ 00172 1 9 10 0.0004 1
£to vironfllffilaJt.ilu Di \'ttl Joru l ,, 115 O.o79S I 8 30 o.oon 1 7 6(1 o.oo27 1 Jle'II(Q.Iion ofll.arrler bland~ I 5 ~ O.OS:u 1 I 2160 O.H\1 I ::; 2880 o.1319 1 HerniYc.y ~tim I 8 20 o 0118 1 9 JO 00019 1 g 30 0.0013 1
Exanine Smlaimbuity ofFish erie;! I ::; 250 0.14'19 1 4 540 0.103.5 1 6 120 O<IJS4 I
Rc:le;lfclUrJ8 lmp:l¢t<l of Climale C11a:tl8e I "' 40 o 02!<i I ;; JO&:i 01070 1 s ~0 00219 1 I-' MQniC(WillB Natur&l ,\aenualiJn of Re~xhu.l Q~ I 2 .500 0.2951! I .5 IW 0.03-1.5 1 2 .5760 0.2639 I en N I I I I
B&iiw EquiJIIlc:nlforResidw.l Oil Clean Up I I 220 o 4230 1 .1 28&3 0.6956 ! 6 20 0 0023 1 £duca1ioo a:nd Ttainio8 1 4 4.5 0.0865 1 4 120 0.0289 1 2 ~0 0.0566 1 E:llabli!h Mooitain,g Sy:lle~u I 2 110 0211.5 1 s .\0 0.0096 1 .5 40 o.00·\7 1
I.CJsi<~>=-•1 lnl<nn41l:ln Slwill8 l 3 ~ 0.17JO 1 2 720 0.1739 1 I 7680 <L~67 1
Slrea.mline Rcwvecy fm)j Pro~ure I 6 15 0.0288 1 3 360 0.0869 1 3 160 0 OISB 1 Funds fct" lncreA-!ed Community }.f~:ings,'l'Jograms I 7 10 0.0192 1 6 10 O.OOH 1 4 80 000~ I
J':!ri;sb Adrn.ini:llratil<e Cap!city I 5 :!<) 0.05"16 1 "' JO 0.0024 1 7 10 o <IJil I j
Funding Ia f',ovenm1<;:ni I 2 4() 02285 l 2 30 02.5()0 1 2 180 0.1800 1
Oulreacll ProgrUI13to Prcmote Tourism/rl;!bcrica l 3 30 0.1714 1 5 10 00833 5 10 o 0100 1
Eron.lnti:; Aqu&<:<~ltute Stud~ l 4 ~~ O.OSS'I 1 4 15 0.1250 I 720 0.72(1J 1
Stjpro:l for lluiioc-.H Adjwiru.C1Jito New Erll'iW ltc-guht~Jtl .5 10 (J.0.571 l 3 2.5 02083 3 60 o.()5oo I J.~~ f(l" Small llu<ine<!.l ~elcJJ-JIC!It I I 80 0.4571 1 I ,t(J .0.3333 ·i 30 0.0300 l
I I I Community Oil tread! and SllJ1la1 Programs I I 64:8 0.4544 I 6 20 0.0990 2 %0 03855 I
l:lolncr Edu:att:.tJI'ropn:3 I 6 !<i 0025:2 1 3 3(1 0.1~5 I %0 0.3855 1 l'llblic lle4lfo liallill8 &~~d 1hJTYcilhn:e ll1.1"1"'1e I ·I !OK O.<J7.57 I .5 22 0.1089 6 60 0.02·10 1
li{ICieal JmpmveCap&citylor!'llbli; lie&ilh l'wgn.m• l 5 71 O.OSGI I 7 15 0.0742 •7 20 O.OO&J I
lntrca.1C Mc:o!a.l Hcaltb C~ity I 3 216 0.1514 1 I 4S 02227 g 10 o 0040 1
Populatim and Dmloga.pbiw Stu:lk! I 1 12 O.((IS' I 4 24 O.lJSS 3 240 0(1963 1
L&n,gll~<e 'f fa:tll!Mioo I 8 10 0.0070 1 8 10 0.(1.\9.5 " 120 o.t~t81 1 lmJirove Mocli:al Servl:e, I 2 32-i 0.2272 1 2 36 0.1782 .5 120 O.OiSi I
Jl,AW.DATA
SCJENC£ AND 'f £CHNICAL JU:l'JU:SENr A: riVES !OXI'!ORT GltOUI' I F:<JH!rt I~ I E•11crt 17 I E'Jlert 18
Objmlvt<~ I R&n~ RAw '>'r'cisld I Rank RAw Wcighj I Rank RAw Weisld I Fn¥ironmem&l I 2 200 (I 0896 I l 4SO 0.7758 1 1 225 06250 1 Loj;i!lli:al Capl.city for Di>Wb'ltespooile I 3 20 o.oos9 1 ~ 10 0.0112 1 3 50 0.13118 1 Re;!;i<W~&li!C(I'I(micSiabi~ly I 4 10 0.0044 1 3 :30 0.0512 l 2 75 O.W&l I So;.ieallm~s I l 2(100 o 8968 1 2 90 0,1551. 1 4 10 o 0211 1
I I 1 H.tlltovalud Di!lp<>lal of Jtc:3idwl Oil l 1 45000 088U 1 l ({.0 0.326? 1 1 2400 orou 1 CY!'~ S~nJl Jl~.,r&ti(l!l I 5 ro 0.0011 1 2 660 0~267 l .t 120 O.OiOO I Cmse.,ci<)n<:d'Q)'!l11;< .1~13 I 6 20 o O<J:l3 I 8 30 0.0148 1 :3 120 o 0400 t
Fnviroomema!RiYcr Dh=iom I 8 lO 0.0001 l ' 160 0.0792 1 6 20 0.0066 1 R~-tiooofflurlu hluldl l 7 20 0.0003 1 1 .to <L0198 t 5 60 0.0200 1 1b'lliv<-Y l'r~tim I 9 10 (U)001 I 9 10 o.oow 1 9 !0 0.0033 1 Examine Swn.imbility of Fisheries I 4 ISO 0.0035 1 6 80 003'96 1 2 240 o 0801 1
1-' Jlt:!eatd!ius Impact! ofC1ima1o Dl!fl~ I 3 900 o.om 1 3 220 01089 1 7 15 0.0050 1 en w Moo~oon,g Nil ural AaC1111Mioo of JWdual Oil I 2 .uoo o.o887 1 4 160 0.0?92 1 8 10 0.0033 1
i I 1 DMio-. Equipmmj fa RCIIi:hLll Oil C1:an Up I 3 12000 o.o089 1 ? 10 00081 1 2 100 0.1724 1 lli.catioo ud Traioin,g I 2 120000 0.0899 1 6 30 0.0245 1 4 50 0.08(o2 1 l'.<nllli~ M()Oiit~ns Sy:ll~n I I 1200000 0.8999 1 5 60 0.0491 1 I 300 0.5172 1
Logis~ko&l Jnfonnatirn Shu:iii!J I 4 1200 0 0080 1 I 400 O.J278 1 3 100 0.1724 1 Slream]inc Recovery Fmd Pro.-.cdorc I {; :!>:) 0.0030 I 3 180 OJ4?5 1 ~ 10 0 0172 I J'Wldl for loorwed C(JIIIfltunity 1fectin,g~'l'r()8t&Otl I 7 10 o.oooo 1 2 360 02950 l 6 10 (1.0172 1 f>arilh AdminisCn.lh'<: Ca.p.wily l 5 lXI o.oooo 1 •I 180 0.1475 1 7 10 0.0172 1
I ! I l'Wldilll! for Ga l'ffllttlt'll1 I 5 JO 0.0243 1 1 21:)0 0.6250 1 2 JOO 04J(o6 1 Qu~ I'rq;nm1., l'rornQir: Tg,ai!iln.'l'i•hori~ I 3 :;.:, o.o731 1 ~ 50 O.IOil 1 3 20 0.0!133 1
r:c.on..1mic 1\q\UCllhue St•-1):, l I :300 0.7317 I 2 100 02083 1 I 100 o 4166 1 Stipmd for ~siOCIIs Aqjwlmmt .I.J New Enviro Regul.\1~n 4 JO 0.0243 I 5 lO 0 0/XI8 l .~ 10 o 0416 I Loaru for S(Jia.IJ flwint:i! DeveJopntetd I 2 00 0.1·163 I ~ 20 0.0~16 1 ~ 10 O.Otl6 I
I I I C.oounuoi~· Ouir=b Mid Suppm Programa I 4 640 0.0017 1 5 60 o o355 1 3 40 0.1025 1 Bols~cr &location Progr8ill;l I 6 40 0 OO:H I 4 120 0.0710 1 4 ~0 o O?(f) 1 J'ubtic lfthb Staffing M~d Surveilh.oce Ll:!.tawe I 2 32000 o.owo 1 2 360 0.2130 1 2 60 0.1538 t
S~eb.l lmprQ\'C Capa<>lly f{l- 1'1•111); ll~a.lO. l'rQSr&lll" I J 6400 o o 178 1 3 :360 02130 1 6 20 o 0512 1
I=e-4.1c MemaJ lbJG:I C~il)' I I 320000 0 S9J7 I I 720 0.4260 1 I ISO 04615 1 l'opllli!Aior• an:~ Detn(J8ra:fil~ Stu:li~ I 8 ](I 000001 6 40 0.0236 1 7 20 0.0512 1 ]Mill"~ Tran•l>1:1.~n l 7 10 0.0000 1 1 20 0.0118 1 8 10 O.o256 I 1mJW<J\'C M«<ai S<:rVl;c;:s I 5 160 oooo4 1 8 10 0.0059 1 5 30 0 07W 1
Jt,\WDATA
SCIENCE AND TECIU.lJC~\L REPRESENT ATlVES EXPERT GROUP 1 f.>.J•erllll
Objecth·"" I )! ... ).; naw We\shll Envirmmemal l 1 1»:1 O.M2B I Logisl:i.:al Capi.;ity for ])jsa;;a Rcspmse I 3 30 01071 1 Jtcsiooa.l Eooooolic Stability I 2 60 02142 1
S<.>Oi~1DIJI'IC1~ I ~ ]() 0.0357 1
I Remava.hll<i 1:>i:~~~~a1 gflle-~>.1w.1 QlJ I j; 2(J O.OtOJ; I Cypreaa Swamp Rt:!tor&1iro I 4 4Q OOS16 1
Cooslrn:ixm ofO)~Iel Rccfu I 3 ro o.l6n 1 flll'it.:IOotffi!alltivct Dhoet>~iora I 1 I~ 0~6S I
R<=!l(llll.tX.n of ll,vri<" l•lm<l• l 5 4Q O.OJ; 16 1
Heroiv~ Pmoealiro I 9 10 0.0204 1 Examine Smwmb~ity of Fisheries 1 2 ro O.J6n I l~ingbnp!!:ti ofC!irna1e C'hulge l 6 4(l 0.0816 1
!--> Moo~c:w-ina N#IIT&1 /l.llo:>t~~&lion or .Resi!ha&1 Q~ I 7 20 O.OtOJ; 1 en +::> 1 l
Baolic fquipmall fc.- lte.~ldwl Oil Clt.ul Up I 5 20 <t<IS(J9 I fdu::o1iioo at~d 'fraioios I 3 ,1() 0.11~ 1 1'.1t.bli"" Mmltori"8 System I 1 &:) 0.3178
Logisl:i.:al lofonna.tim Sl:wiD8 I 2 4Q 0.1739
SII'Ol:l•tline llecoVC1Y Fund l'roctdu.to I 6 20 0 _(18(19
J-"undl fc.- ln.::ca:~~ Cmnmunity M~D8s'1'rogtarn.!l I 4 2() 0.01;69
P....Uh Administrath..; C&~ty I 1 10 o 04:34 I
I Fundios fc.- Go\'Cmratud I 3 20 0.1176 l Ou~ l'rog~111;1 l'roo10il: T<JIIri,..lfi'J.<I...-i~ I l &:) O.·t?OS l
f..:.<ln..,..U, Aq1=uhure S ludk! I 2 4Q 0.2352 1
Slipendfor Busines~ Aqjmvn~t to Nc:w EI!Viro Reguht)tn s 10 o OS8!1 I Lollftl fc.- Small Bwioe>U De~lopoiMI I 4 20 0.1176 1
I Canm11nity Ou~ andSIIJIII<W1 J'rt)~L1 I 4 4Q o 12!(1 1
Bolsle:r Edooatxm Prognw I 6 20 0.0625 I l'ublic liea.ht• Staffins a11d SurvdiJ.u,~ Daia.l:we ! 5 4(l o.mo 1
1)(10i$l 1m1.-o~ Cq~ 1<.- l'uhli:; 1 bill l'rogram. l l &:) 0.2500 1 1n=&c<e Mem.l 1 bl" Cqncity I 3 4Q o 12!(1 1 Popul&1im an:! Dcmogr&Jili; Stu.:liea I 7 10 00312 1 WSIJ* ·rwal11.ion I 8 10 0_(1312 l
)mJI'o'..; Med>:.J S<"V~ I 2 ro 0.2500 I
RAWDAT.\
COMMUNITY llASI:JJ REI'RBSE!ff 1\TIVf.S I':Xl'ERT GROlJI'
E1J!erll0 I Ex penH 1 r.xpen n O~jedlvto:S I Ramk Raw Wcigbt I Rani: Raw Wcigllll Rank RAw Wei81d 1 l:llvitoollltfdal I I I (I O.OSOO I 1 50(1 0.7936 ' I 66 0.«155 1 Lagi:llnl Ca~ity fQr DiM.111:2" Re:!jlm~ I 2 .j(J 0.2000 1 4 ](I 0.0158 l 2 22 0.2018 1 Regimal Ecoocmic Sl&bilily I 3 5IJ 02500 l :3 40 o 0634 1 3 11 o 1009 1 S«:ielilllm~s I 4 100 0.5000 1 2 8(1 0.1269 1 4 10 0.00!7 1
I I Retna~al81ld Ui:lp(!lal afR~JidUil Oi1 I 8 to 0.02t9 l 9 to 0.0(113 1 7 13 0.0285
Cypre<l S""""l' R~t1rati1U1 I I too 02197 l 3 160 0 I t76 1 _I _192 0.4210
fmiroontttdal ltiYtt Di...wi:Jw I 2 100 01197 1 I 500 OJ67(i 1 3 48 0.1050
l~cr&iionofl:lurinlila•xb l 7 1S 0.(629 l ·I 80 0.051!8 1 s 22 0.~182
lkotiv(ll)' ""'-tiQO I 4 ro 0.1318 1 6 40 0.0294 l 2 96 02105
Exa.mincSullil.imbllily ofFish~ l 5 ((l 0.1318 I 7 20 0.0147 I 6 20 0.0438 l~ioghup~ ofC!imato Chul&-e I 6 :w O.o659 I 2 YJO 0.1616 1 9 to 0.0219
M<W~~<Jril18 N&~ur&l A•enu..C<W~ of ~d1Joll 0~ I 9 to 0.0219 1 8 to 0.0(113 1 8 11 0.0241 f-' O'l I I ' LT1
l:lui¢ I::quipnltt•l f« J~id.W Oil Cleul Up I 100 0.!3t'9 I 7 10 0_0250 1 I 4 45 (10576
l:du:.tt:ion 81ld Trainill8 l 5 25 o.os.n t 1 160 0.4000 l 3 225 0.21!8-1
r:1v.b~sb Mrn~mng Sy:~~e~n I 3 4() 0,1355 I 5 20 0 .OSIJO I 2 225 02884
l.ogiSI~ InfonnMim Sh&rill8 I 4 4() o,l355 1 4 40 0,1000 1 I 225 02884
Slrt81l!lioe llt.XIY~' t'U!Id l'roeedure l 6 20 0.0677 1 6 10 o 0250 1 6 20 0_0256
fun~l• fc:r ln:rease~IC11mmuni!y M~ng .. •l'rog,..n.! l 7 10 o.ill38 I s so 02000 l 7 ](I 0.0128
_l"ari!h A<lmini:~tn.th"' C~ty I 2 ro 0,2033 I 2 80 02000 1 5 30 0,0384
I I t'U!Iding f« Go Yt1110lttd I I 90 0.3-161 1 I 20 0.1333 .I 5 ](I o.0085 1 011~• Progr~n!! "' l'r<ITIQ!o T Quri:.,nll'i!hori~ I 2 70 0.2692 I 3 10 0.1666 1 3 50 o.~1Z7 1
EcooonU; r\q~~&o:-ullurc Studle;J I 4 4() 0.1538 1 5 10 0.1666 1 I 1000 o 8547 1 Stipend for 11-u:iirlt:l! Adjlllb\1(111 tl New En~iro Rtg~Jluito 5 10 0.<138-1 I 4 10 0,1666 l 4 JO 0.0085 1 I.Qilll tc:r Sman ll11!inetl IJ..,..,l~menl I 3 50 0.1923 1 2 10 0.1666 1 2 IOO 0.085\ 1
I I I I Cooummity Oulrcacll Uld Supprnl'rogra.ms I 4 ((l 0.1395 l > so 0.0956 1 5 150 0.0038 I
l:lal,.ttl:du:ationl'rograrn.! l 5 40 o <1930 I l 500 05980 1 6 15 0 0019 1 l'ul•li~> lb11h Slall'i:ns &lid Swv<:OIW.~ IAIA\>1.1~ I 2 90 0.2093 I ~ ·10 0.047l! 1 ·I ISOO 0.0382 1
Enviroomcmal River DiYt:111i1ru I 2 3((10 a.279? I 2 1::xl 0.1411 1 5 1.5(10 (I (l.f39
R~t<ntiou of &triet l:!laro~ I 1 40 0.0031 1 6 40 o.ono 1 6 1.51JO 0.0439
Uerbivcry l're\...,ti\WI I 9 10 o.OOQ1 I 9 10 0.0117 1 9 10 0.0002
E~~neSu1t1.inabillty ()J'Fi~·~ I 6 so 0,0062 1 5 40 0.0470 1 3 3000 0 0879
Re~Jl8 Impac.tl mC!ima.1eO.a.nge I 5 320 0.0248 l 7 20 0.0235 1 ? 1000 oo:m 1 1-' O'l Monitc.ing N&!utal Aletllllilian of R~idUll Oil I 4 .wo O.Cr310 I 8 2(1 O.Cr235 .I 8 100 0.0029 1 O'l
l I l I &1~ EquiJm~l for Re;~idml 0~ Clea.n Up l 4 90 0.0466 1 5 10 0.0222 l I 3ffl0 o.76B 1 Education and Training I 3 135 0.0699 I 2 12(1 0.26ift 1 2 ?20 0.1528 I
.E:IIabliJh MooitcrirJ& Sy:lletn I 6 30 O.OJ.5S I .. 20 0.04·1-1 1 s 4() o.oos.1 1 l.()gi.<lnl lnlalt\'llion Slu.rif18 I 1 121.5 0.6295 1 1 2·1() o.s:;r, 1 3 2.0 0.0.509 1
Stre4Dilioe Recovery Fund Pro(.C(jjue I 2 405 O.XI98 I 3 40 0.0888 l 4 80 00169 1
Funds fa ln~ C.ommunity:Med:iogs1Program.1 I 1 10 O.OJ51 l 1 1(1 0.0222 1 1 10 0.0021 I
l'ui:4! Adutini:llr&1iye Capacity I 5 45 0.02n 1 6 10 0.0222 1 6 20 o.o~12 1 I i I
Funding fa Gcm:mm~1 I 3 60 0.0937 1 3 60 0.0937 l 2 .500 0,1845 1
Outreac.h Programs to Pr001ole Tolllismil'isbe:rie;~ I 1 J(o(l 0.5625 I 2 1&::1 0.2812 1 3 ISO (10553 1
l:i.!ofJOfJLic AqUiCUbute Stu:lie-J I ·1 30 0.~~68 1 5 10 O.OB6 I I 2000 0.73110
S1ipeod ll:lr fl.u1in~ ;.<lill11rnen11~ :New Envir<J Regub.1~m s 10 0.0156 1 ·I 30 0.~168 l 5 10 0.0036
Loem fa Small BruinC;I;!I DevcloJm~t I 2 180 0.2812 1 I Jro 0 . .5625 l 4 50 00184
I I I Comrnuoity Q,fl'e!dl and Suwm l'tograru~ I 6 60 o.02S6 1 s 10 O.OISJ I 8 10 0.0016
Public lb.Jfl Slilffill8 and SurYaJJao;e Dat&bi;Je I 1 I ZOO os139 1 5 20 0.0303 1 3 .500 ooros Sooetil lmptove Capacity fa !'ubi~ lb.ldJ l'togram~ I 2 ((JO 0.25(8 I 2 1((1 0.2424 t 2 2500 04(129
ln=e M~al Htaltb C~l)' I ,f 160 o.06~ I ·I .fO tt0606 1 4 120 0.0193
L a.ng t1 S@C: T ra.ru la1 i1 o I 8 10 0.0042 1 ? 10 o,OIS! 1 5 40 00064
JruproveMedi.:al Sen'~ 1 3 200 0.0856 1 I 32(1 0.4tt4tt I 1 3000 0.4834
RAW DATA
COMMUNJT Y n,o,.;sr:D Rl; !'RP.; f;Nf A Tl.Vf.S EX !'ERT (lROV I' Ullotril6 I £ipotrl17 I :U1~otrllll I
Objtdlves l lWil luw Wei,ght I Jw& Raw Wd81tC! lWil luw Wd,gld I
F:nvirmm"""&l I I 100 o.?6112 I 2 .,o 0.19:1-t I I 60 o .. \615 1
Logistal C&pQCily for Disa:st:r Re!prnsc l 3 10 0.0769 I I 120 05714 1 4 10 o 0769 1 Rcgirna.l Ecrncmic Sc&bilily I 2 10 O.O?W I 3 40 0.19:14 I 3 30 0.2Wl 1 Sociml bnpact~ I 4 10 o.0769 1 ~ 10 O.O.t76 I 2 30 0.2307 1
I l I R.em<TY&l Uld J)j"''(ll&l ofl!e1xhl!-1 011 I 7 ::xl O.o\00 1 1 720 O.H.Sl! t 8 10 0.0103 1
Cypre33 Swamp Res~raeirn l 2 ro o 1600 1 5 90 0.0557 1 I 320 0.329!1 1
Cortslru:Hon ofOyJiet Rt"Cfi I I 160 OJ2W I 6 90 0.0557 1 2 320 0.3298 1 Enviroofllttda.l RiYtt Di\tt!lionl I 6 "" 0.0800 1 3 ISO 0.1 ll.t 1 3 160 0.16491
Re!I<F&ll_1n <:tfll.uric:r 1~.,.;1~ I 4 60 o 1200 1 2 :360 02229 1 5 40 o 0412 1
lbbivay l'revrnlioo I 9 10 0.0200 I 4 90 0.0557 1 9 10 00103 I E.u.nW!e Swl!i~bijity ofFilllffiel I 3 XI 0.1400 1 8 30 o 0185 1 4 80 o .0824 t R-.vcltins lmJ"'c"' <:tf Clim~11:: C'hUtse I 5 ·\5 0.0900 1 9 10 0.0051 1 6 20 0.0206
Sripc1ld for I:I'U:Iioe:!s Aqjwtu~ll:t ~ £r1YitO R~'llhlitrt I 6() O .. tBOO I 2 120 0.17111 5 )() o.oo86 1 l.(l&nl f(lr Small 11-!t.linel.'l D~l~mCI!ll I '2 :J>:l 02400 1 l 4110 0.7164 I 600 05172 1
I I Coo1.11111rthy Outrt-aclt a.nd SuwM l'rogra.nu I 3 2(1 00833 1 1 320 0.4507 ? 10 0.0192 1 ll<:tl~c:r &hr~xm l'rQsnt.n'" I 2 "" 0.1666 1 6 30 O.O•t22 6 20 0.0381 I
Public l:k&lft Sta.ffing a.nd Survc~la.nc.c Dll.t&bo!c l 5 15 o 0625 1 4 60 00815 2 90 o 17:J>J 1
So;.iC'Ial lmp'Q~ Capacity fCl' l'llblic l:k&ldl Progr&m.3 I 1 12(1 05000 l 2 160 0.2253 3 90 01?301 !n..~CaJe Metda.l lb.! Ill C;tpacily I
,, 15 0.0625 1 3 so 0.1126 1 1110 03\61 1 l'qt\ll&'lim ..,.,i D<m<JBilo}il); Sin.i~ I 6 10 0.0416 1 5 30 O.O-t22 " 60 0.1153 l
La:nguagc Tnruhlion I 7 10 00416 1 7 20 002111 g 10 o 0192 1
lrtlJWO\'e Meodkal SttV~ I 8 10 O.OH6 I 8 10 0.0140 5 60 OJ153 I
RAW DATA
COMMUNI! Y BASED REl'RB'!ENf A TIVFS EXPERT GROUP
r~rt21l 1 F.>: pert Jl()
Objtdh't\! I R-""k R...., w•1 R4llk R•w w• £tJ viromuenlal I 1 60 0 . .5217 1 I 260 o .. IJ26 1 I..<J!iSiiaLI C.q~acity fof Di!.,.,lo::f R¢'111Qil'l~ I 4 10 0,01164 1 4 10 O.CI277
RC(iooal.&x«<~ Slabilily I 2 30 o.zros 1 3 100 O.L~7
SacietallntJ m:ti l 3 IS o.l30·1 1 2 260 0..1126 l I
RmDV&I and DnpoS&l ofRC3idual Oil I 9 10 0.0118 1 ~ 13(1 0.0557
Cypr~• Swa>q1 llO!lor!.IJo21 I 2 Z70 0.319S l 2 511S 0,2507
Cm~tnlcti<m of Oy Sle<r R¢~ I 6 4S 0.0532 1 s 100 0008
l.aoJJuSI.ll' Tn.mhli<m I s 10 0.0117 1 8 10 0.0070
luLprove M ttld Servicn; l ,, 120 0.1411 I 1 3SO 0.2a!2
References
Adams, C., Hernandez, E., Cato, J., The economic significance of the Gulf of Mexico related to population, income, employment, minerals, fisheries, and shipping. Ocean & Coastal Management 47 (2004) 565-580
Aguilera, F., Mendez, J., Pasaro, E., Laffon, B. (2010) Review ofhealth effects to spilled oils on human health. J. applied toxicology 30:291-301.
Alemi, F., Gustafson, D. (2007). Decision Analysis for Healthcare Managers. Health Administration Press. Chapters 1, 2, 3, 5, 6, 10, 11.
American Red Cross. (2003). Disaster Services Connection. Change in the Official Definition of ""Disaster"" and the Addition of a Definition of Community Emergency". #182.
Austin, D., B. Carriker, T. McGuire, J. Pratt, T. Priest, and A. G.Pulsipher. 2004. History of the offshore oil and gas industry in southern Louisiana: Interim report; Volume I: Papers on the evolving offshore industry. U.S. Dept. of the Interior, Minerals Management Service, GulfofMexico OCS Region, New Orleans, LA. OCS Study MMS 2004-049.98 pp.
Austin, D. E., T. Priest, L. Penney, J. Pratt, A. G. Pulsipher, J. Abel and J. Taylor. 2008. History ofthe offshore oil and gas industry in southern Louisiana. Volume I: Papers on the evolving offshore industry. U.S. Dept. ofthe Interior, Minerals Management Service, Gulf of Mexico OCS Region, New Orleans, LA. OCS Study MMS 2008-042. 264 pp.
Bankoff, G., Frerks, D. Hilhorst (eds.) (2003). Mapping Vulnerability: Disasters, Development and People. Earthscan Publishing.
Barron, F., Edwards, W. (1994). SMART and SMARTER: Improved Simple Methods for Multiattribute Utility Measurement. Organizational Behavior and Human Decision Processes. Vol. 60 pp. 306-325.
Batker, D., de la Torre, I., Costanza, R., Swedeen, P., Day, J., Boumans, R., Bagstad, K. Gaining Ground: Wetlands, Hurricanes and the Economy: The Value of Restoring the Mississippi River Delta. Earth Economics, 2010.
169
Blanchard, W. Guide to Emergency Management and Related Terms, Definitions, Concepts, Acronyms, Organizations, Programs, Guidance, Executive Orders & Legislation. P.914. 10/22/2008.
Bloom, D., Canning, D. (2008). Population Health and Economic Development. Commision on Growth and Development. Working Paper No. 24
Bolstad, Erika. "Science world skeptical at oil spill's disappearing act- Gulf Oil Spill". MiarniHerald.com. http:/ /www.miamiherald.com/20 10/08/04/17 61951/science-worldskeptical-at-oils.html. Retrieved 2010-09-05.
Borcherding, K., Eppel, T., von Winterfeldt, D. (1991). Comparison of Weighting Judgments in Multiattribute Utility Measurement. Management Science Vol. 37 No. 12 pp. 1603-1619.
British Petroleum. BP press release. http:/ /www.bp.com/genericarticle.do?categoryld=2012968&contentld=7061778
Brown, B. ( 1968). Delphi Process: A Methodology Used for the Elicitation of Opinions of Experts. Santa Monica, CA: RAND Corporation, 1968. http://www.rand.org/pubs/papers/P3925. Bureau of Ocean Energy Management, Regulation and Enforcement. (2011). BOEMRE OCS production 2009. http://www.boemre.gov/stats/PDFs/OCSProductionTemplate2009.pdf
Bureau of Land Management. (1982). Ixtoc impact assessment.
Cato, J. (2009). Gulf of Mexico Origin, Waters, and Biota. Volume 2, Ocean and Coastal Economy. Texas A&M Press
CBS News. U.S. Says 75% of oil gone, skeptics remain. http://www.cbsnews.com/stories/20 1 0/08/04/nationallmain67 41897 .shtml
Centre ofDocumentation, Research and Experimentation on Accidental Water Pollution. (2011). CEDRE Glossary. http:/ /www.cedre.fr/ en/ glossary. php
Chandra, A, Acosta, J., Stem, S., Uscher-Pines, L., Williams M., Yeung, D., Garnett, J., Meredith, L. (2011). Building Community Resilience to Disasters: A Way Forward to Enhance National Health Security. Santa Monica, CA: RAND Corporation. http:/ /www.rand.org/pubs/technical_reports/TR915 ..
Chocholik, J.K., Bouchard, S.E., Tan, J.K.H., & Ostrow, D.N. (1999). The determination of relevant goals and criteria used to select an automated patient care information system: A Delphi approach. Journal of American Medical Informatics Association, 6(3), 219-233.
Corder, G., & Foreman, D. (2009). Nonparametric statistics for non-statisticians: A step-by-step approach. P.99-100.
Corn, M.L., & Copeland, C. (2010). The Deepwater Horizon Oil Spill: Coastal Wetland and Wildlife Impacts and Response. Congressional Research Service. 7-5700.
Costanza, R., Daly, H. (1987). Toward and Ecological Economics. Ecological Modeling, 38: pp. 1-7.
Costanza, R., Farber, S.C., Maxwell, J. (1989). The Valuation and Management of Wetland Ecosystems. Ecological Economics 1:335-361.
Couvillion, B.R., Barras, J.A., Steyer, G.D., Sleavin, William, Fischer, Michelle, Beck, Holly, Trahan, Nadine, Griffin, Brad, and Heckman, David, 2011, Land area change in coastal Louisiana from 1932 to 2010: U.S. Geological Survey Scientific Investigations Map 3164, scale 1:265,000, 12 p. pamphlet.
Crone, T., Tolstoy, M. (2010). Magnitude of the 2010 Gulf of Mexico Oil Leak. Science. Vol. 330 no. 6004 p. 634
Day, J.W.; Martin, J.P.; Cardoch, L.; Templet, P.R. (1997) System functioning as a basis for sustainable management of deltaic ecosystems. Coastal Management. 25, 115-153.
Department of Energy. (2011). Impacts oflncreased Access to Oil and Natural Gas Resources in the Lower 48 Federal Outer Continental Shelf. Retrieved from: http://www.eia.doe.gov/oiaf/aeo/otheranalysis/ongr.html
Department oflnterior. (2010). DOl issues directive to guide safe six-month moratorium on deepwater drilling. http:/ /www.doi.gov /news/pressreleases!lnterior-Issues-Directive-toGuide-Safe-Six-Month-Moratorium-on-Deepwater-Drilling.cfm).
Department of the Interior. (2010). Summary of Preliminary report from the Flow Rate Technical Group. May 2010.
Department ofthe Interior. (2011). Strategic Scientific Working Group Library. Retreived from: http://www.strategicsciencesworkinggroup.com/omeka-1.2.1/Dewan, Shalia. The Oil Spill's Money Squeeze. New York Times 9112/2010
Diakoulaki, D., Mavrotas, G., Papayannakis, L. (1994). Determining Objective Weights in Multiple Criteria Problems: The Critic Method. Computer Ops Res. Vol.22 No.7 pp.763-770.
Dodgson, J., M. Spackman, A. Pearman and L. Phillips (2000). DTLR multi-criteria analysis manual.
171
Dooley, K. (1996), "A Nominal Definition of Complex Adaptive Systems," The Chaos Network, 8(1): 2-3.
Edwards, W. How to Use Multiattribute Utility Measurement for Social Decision Making. (1977). IEEE Transactions on Systems, Man, and Cybernetics. Vol 7 No.5 pp. 326-340.
The Economist. BP and the Oil Spill: The oil well and the damage done. 6/17/2010
Federal Emergency Management Agency. Developing the Mitigation Plan: Identifying Mitigation Actions and Implementation Strategies (FEMA 386-3). Washington, DC: FEMA, April, 2003. http://www.fema.gov/library/viewRecord.do?id=1886
Flournoy, A. Three Meta-Lessons Government and Industry Should Learn From The BP Deepwater Horizon Disaster and Why They Will Not. Environmental Affairs Law Review. Vol. 38, iss 2. http://lawdigitalcommons.bc.edu/ealr/vol38/iss2/4
Folke, C., Hahn, T., Olsson, P., Norberg, J. (2005). Adaptive Governance of Social-Ecological Systems. Annual Review ofEnvironmental Resource. 30: 441-73
Fos, P.J., & Zuniga, M.A. (1999). Assessment of primary health care access status: an analytic technique for decision-making. Health Care Management Science, 2, 229-238.
Fos, P.J., Miller, D., Amy, B., Zuniga, M.A. (2004). Combining the Benefits of Decision Science and Financial Analysis in Public Health Management: A Country-Specific Budgeting and Planning Model. J Pubic Health Management Practice 1 0(5) p 406-412.
Forman, E., Gass, S. (2001). The Analytic Hierarchy Process-An Exposition. Operational Research. Vol. 49 No. 4 pp 469-486.
Fullop, J. Introduction to Decision Making. Laboratory of Operations and Decision Systems, Computer and Automation Institute. Hungarian Academy of Sciences.
Goldenberg, S. (2010). BP Oil Spill Ruined My Life, Says Louisiana Shrimp King. Guardian.co.uk, 6111/2010. Retrieved from: http://www.guardian.co.uk/environment/2010/jun/11/bp-oil-spill-shrimp-king
Goldstein, B. D., Osofsky, H. J., & Lichtveld, M. Y. (2011). The Gulf Oil Spill. New England Journal ofMedicine, 364(14), 1334-1348. doi: 10.1056/NEJMra1007197
Green, S., Moss, G.W. (1998). Value Management and Post-Occupancy Evaluation: Closing the Loop. Facilities. Vol. 16 No. Yz pp.34. Retrieved from:
Gregory, R., Keeney, R., von Winterfeldt, D. (1992). Adapting the Environmental Impact Statement Process to Inform Decisionmak:ers. Vol. 11 No. 1. Journal ofPolicy Analysis and Management. pp. 58-75. Retrieved from: http://www.jstor.org/stable/3325132
Grey, E. (2010). Tulane researcher finds evidence of oil in the Gulf food chain. WWL New Orleans. Retrieved from: http://www.youtube.com/watch?v=pPqSgLORD8U
Guardian. BP oil spill tower fails. http://www.guardian.co.uk/environment/2010/may/09/bpoil-spill-tower-fails
Hammond, J, Keeney, R., Raiffa, H. (1998). The Hidden traps in Decision Making. Harvard Business Review. September/October 1998.
Haralambopoulous, D.A., Polatidis, H. (2003). Renewable Energy Projects: Structuring a Multi-Criteria Group Decision-Making Framework. Renewable Energy. Vol. 28 pp. 961-973.
Hasson F., Keeney S. & McKenna H. (2000). Research guidelines for the Delphi survey technique. Journal of Advanced Nursing 32, 1008-1015.
Hoffman, R., Shadbolt, N., Burton, A., Klein, G. (1995). Eliciting Knowledge from Experts: A Methodological Analysis. Organizational Behavior and Human Decision Processes. Vol. 62 No.2 pp 129-158.
Houck, 0. (2010). Worst Case Scenario and the Deepwater Horizon Blowout: There Ought To Be a Law. Tulane Environmental Law Journal. Winter, pp. 33-39
Hsu, C., Sandford, B. (2007). Minimizing Non-Response in the Delphi Process: How to Respond to Non-Response. Practical Assessment, Research & Evaluation. Vol. 12, No.
17.
Jarvis, J. (2010). Deep Water Horizon Oil Spill: Impacts on Coastal Plant Communities. Marine Science, The Richard Stockton College ofNew Jersey.
Jemelov, A., Linden, 0. Ixtoc 1: A case study ofthe world's largest oil spill. Ambio Vol. 10,
No.6, The Caribbean (1981), pp. 299-306. http://www.scribd.com/doc/32237183/Ixtoc-1-a-Case-Study-of-the-World-s-Largest-Oil-Spill
Jones, J., Hunter, D. (1995). Qualitative Research: Consensus methods for medical and health services research. BMJ 1995;311:376-380 (Published 5 August 1995)
173
---- ~--~~-~- -----~~-----
Keirn, M. (2008). Building Human Resilience: The role of public health preparedness and response as an adaptation to climate change. American Journal of Preventative Medicine. Vol.35, no.5, pp. 508-516.
Keeney, R. (1982). Decision Analysis: An Overview. Vol. 30 No.5 p.803-838. Operations Research. Retrieved from http:/ /www.j stor.org/stable/17034 7
Keeney, R., von Winterfeldt, D., Eppel, T. (1990). Eliciting Public Values for Complex Decisions. Vol.36, No.9 Management Science. Retrieved from: Http://www.jstor.org/stable/2632353
Keeney, R. & Raiffa, H. (1992). Decision with multiple objectives: Preferences and value tradeoffs (2nd ed.). Cambridge: Cambridge University Press.
Keeny, R., McDaniels, T. (1992). Value-focused Thinking About Strategic Decisions at BC Hydro. Interfaces. Vol. 22, pp. 94-109.
Keeney, R. (1994). Creativity in Decision Making with Value-Focused Thinking. Sloan Management Review. pp.33-41.
Keeney, R. (1996). Value-focused thinking: Identifying decisions opportunities and creating alternatives. European Journal of Operational Research. 92, 537-549.
Keeney, R. (2001). Common Mistakes in Making Value Trade-Offs. Operational Research. Vol. 50 No.6 pp 935-945.
Kerr, R., Kintisch, E., Stokstad, E. (2010). Will the Deepwater Horizon Set a New Standard for Catastrophe? Science. Vol. 328 no. 5979 pp. 674-675.
Kundal, H., Polansky, M. (2003). Measurement of Observer Agreement. Radiology. August pp.
303-308.
Ko, J. Day, J. (2004). Wetlands: Impacts ofEnergy Development in the Mississippi Delta. Encyclopedia of Energy, Vol. 6
Landis, J. R., & Koch, G. G. (1977). The measurement of observer agreement for categorical data., 33(1), 159-174. International Biometric Society. Retrieved from http://www.jstor.org/stable/252931
Lake Pontchartrain Basin Foundation. (2011). History of the Pontchartrain Basin. Retrieved from: http://www.saveourlake.org/basin-history.php
Linstone, H. & Turoff, M. (2002). The Delphi Method: Techniques and Applications Retrieved from: http://is.njit.edu/pubs/delphibook/
174
Louisiana Coastal Wetlands Conservation and Restoration Task Force and the Wetlands Conservation and Restoration Authority. 1998. Coast 2050: Toward a Sustainable Coastal Louisiana. Louisiana Department ofNatural Resources. Baton Rouge, La. 161 p.
Machlis, G., & McNutt, M. (2010). Scenario-Building for the Deepwater Horizon Oil Spill. Science. Vol. 329 no. 5995 pp. 1018-1019
Mendelssohn, I. A., M. W. Hester, C. Sasser, and M. E. Fischel. 1990. The effect of a Louisiana crude oil discharge from a pipeline break on the vegetation of a southeast Louisiana Brackish marsh. Oil and Chemical pollution 7:1-15.
Mendelssohn, I. A., M. W. Hester, and J. M. Hill. 1993. Effects of Oil Spills on Coastal Wetlands and Their Recovery: Year 4 Final Report.
Mulhall M. (2007). Saving rainforests of the sea: An analysis of international efforts to conserve coral reefs Duke Environmental Law and Policy Forum 19:321-351.
New York Times. Map and Estimates of the oil spill in the Gulf of Mexico. http://www.nytimes.com/interactive/2010/05/01/us/20100501-oil-spill-tracker.html
National Oceanic and Atmospheric Administration. (2011). NOAA celebrates 200 years of science, service, and stewardship. Retrieved from: http://celebrating200years.noaa.gov/
NOLA.com. Federal report downplaying drilling moratorium effects is disputed by Mary Landrieu, Vitter. 9/16/2010. http://www.nola.com/news/gulf-oilspill/index.ssf/20 1 0/09/federal_report _that_ drilling_ m.html
Palinkas, L., Pettersen, J., Russell, J., Downs, M. (2004) Ethnic Differences in Symptoms of Post- traumatic Stress after the Exxon Valdez Oil Spill. Pre-hospital and Disaster Medicine. vol. 19, no.1
Public Broadcast System. (2010). Three-quarters of gulf oil spillis accounted for, government says. http://www. pbs.org/newshour/rundown/20 10/0 8/three-quarters-of-gulf-spill-oil-isaccounted-for-government -says.html).
Ramanathan, R. A note on the use of the analytical hierarchy process for environmental impact assessment. Journal of Environmental Management (2001) 63, 27-35
Richardson, N. (2010). Deepwater Horizon and the Patchwork of Oil Spill Liability Law. Resources for the Future. June. pp. 1-6.
Rhodes J , Chan, C., Paxson, C., Rouse, CE., Water, M., Fussell, E. (20 1 0). The impact of hurricane Katrina on the mental and physical health oflow-income parents in New Orleans. Am J Orthopsychiatry. 80(2):237-47.
Roach, J. (2005). Gulf of Mexico "Dead Zone" is Size ofNew Jersey. National Geographic. Retrieved from: http:/ /news.nationalgeographic.com/news/2005/05/0525 _ 050525 _ deadzone.html
Robertson, C. Efforts to Repel Oil Spill Are Described as Chaotic. New York Times Jun 14, 2010.
Robinson, W. (1957). The Statistical Measurement of Agreement. American Sociological Review. Vol. 22, No.1, pp. 17-25.
Roe E. 1998.Taking Complexity Seriously: Policy Analysis, Triangulation and Sustainable Development. Boston (MA): Kluwer Academic Publisher
Rudolph, J.C. (2010). Dead Coral Found Near Site of Oil Spill. New York Times, 11/5/2010. Retrieved from: http://www.nytimes.com/20 1 0/11/06/science/earth/06coral.html
Saaty, T. (2008). Decision Making with the Analytical Hierarchy Process. Int. J. Services Sciences. Vol.l No.1, 83-98.
Saaty, T. (1999). Fundamentals of the Analytic Network Process. ISAHP 1999, Kobe, Japan, August 12-14, 1999.
Schenkman, L. (2010). Gulf Oil Spil: Three Historic Blowouts. Science. Vol. 328 no. 5979 p. 675
Schleifstein, M. (2010). Scientists wary ofBP oil spill's long-term effects on species. The Times-Picayune, 11/10/10. Retrieved from: http://www.nola.com/news/gulf-oil spill/index.ssf/20 1 0/11/scientists _wary_ of_ bp _oil_ spil.html
Shook, G., Fos, P. (1993). An Environmental Health Evaluation Tool for Locating and Assessing Disaster Relief and Refugee Camps. Journal of Environmental Health Vol. 55 No.7. pp. 21-23.
176
Short, J.W., M.R. Lindeberg, P.M. Harris, J.M. Maselko, J.J. Pella, and S.D. Rice. 2004. Estimate of oil persisting on the beaches of Prince William Sound 12 years after the Exxon Valdez oil spill. Environmental Science & Technology 38(1): 19-25.
Siegal, S. (1957). Nonparametric Statistics. The American Statistician. Vol.l1, No.3. pp. 13-19.
Siegel and Castellan. (1988). "Nonparametric Statistics for the Behavioral Sciences" (second edition). New York: McGraw-Hill.
Skulm.oski, G., Hartman, F., Krahn, J. (2007). The Delphi Method for Graduate Research. Journal of Information Technology Education. Vol. 6, pp. 3-21
Slomski, A. (20 1 0). Experts focus on Identifying, Mitigating Potential Health Effects of Gulf Oil Leak. Journal of American Medical Association 304(6):621-624.
Stokstad, E. (2010). Louisiana Begins Controversial Engineering to Ward Off Oil Spill. Science. Vol. 328 no. 5983 pp. 1214-1215
Suarez, B. Lope, V., Perez-Gomez, B. (2005). Acute Health Problems Among Subjects Involved in the Cleanup Operations Following the Prestige Oil Spill in Asturias and Cantabria (Spain). Environmental Research. Vol. 99, Issue 3. pp. 413-424.
Templet, P. H., & Meyer-Arendt, K. J. (1988). Louisiana wetland loss: A regional water management approach to the problem. Environmental Management, 12(2), 181-192. doi: 10.1 007/bfD1873387
Turner, KJ. (2002). Do information professionals use research published from LIS journals? 68th IFLA Council and General Conference. August 18-24, 2002.
United Nations. (1997). Glossary of Environment Statistics, Studies in Methods, Series F, No. 67, United Nations, New York.
United States Army Institute for Water Resources. (2002). Trade-Off Analysis Planning and Procedure Guidebook. April 2002.
United States Geological Survey. (2000). Nutria, Eating Louisiana's Coast. USGS FS-020-00 (Updated 4/20/01).
United States Government Accountability Office (2009). Hurricane Katrina: Barriers to Mental Health Services for Children Persist in Greater New Orleans, Although Federal Grants Are Helping To Address Them. GA0-09-563.
United States Senate Ad Hoc Subcommittee on Disaster Recovery. (2011). Testimony of Kenneth R. Feinberg Administor, Gulf Coast Claims Facility.
177
---- ----- --~--···----- ~·-·---·--
Von Winterfeldt, D. (1980). Structuring Decision Problems for Decision Analysis. Acta Psychologica vol. 45 pp.71-93
Von Winterfeldt, D. (1982). Settling Standards For Offshore Oil Discharges: A Regulatory Decision Analysis. Operations Research. Vol 30 No.5 pp. 867-886.
Von Winterfeldt, D. & Edwards, W. Decision analysis and behavioral research. Cambridge University Press: Cambridge (1986).
Wenstop, F., Knut, S. (2001). Legitimacy and Quality of Multi-Criteria Environmental Policy Analysis: A Meta Analysis of Five MCE Studies in Norway. Journal of Multi-Criteria Decision Analysis. Vol 10 pp. 53-64.
West, J. (1981). Ixtoc I Oil Spill Litigation: Jurisdictional Disputes at the Threshold of Transnational Pollution Responsibility. 16 Tex. Int'l L.J. 531.
World Wildlife Fund. (20 11 ). The Importance of Coral to People. Retrieved: http://www. wor ldwildlife.org/whatlwherewework/ coraltriangle/importance-of-coral.html
Yoe, C. Trade-Off Analysis Planning and Procedures Guidebook. U.S. Army Corps of Engineers. April, 2002.
Zuniga, M., Carillo-Zuniga, G., Ho Seol, Y., Fos, P.J. (2009). Multi-criteria Assessment of County Public Health Capability Disparities. Journal of Health and Human Services Administration. Vol23 No.3 pp. 238-258.