The contents of this report reflect the views of the authors, who are responsible for the facts and the accuracy of the information presented herein. This document is disseminated under the sponsorship of the Department of Transportation, University Transportation Centers Program, in the interest of information exchange. The U.S. Government assumes no liability for the contents or use thereof.

TECHNICAL REPORT DOCUMENTATION PAGE

The Research Team wishes to acknowledge the North Carolina A&T State University Transportation Institute' Urban Transit Institute for their decision to fund this research project as and award us the opportunity to conduct

Particular thanks are due to Melanie Hylton (Undergraduate Student, College of Business - Transportation and Logistics Major), Tony Reames (Undergraduate Student, College of Engineering - Civil Engineering Major), and Hector Williams (Undergraduate Student, College of Business - Transportation and Logistics Major) for their two extraordinary student presentations of the research results given during the 81st Transportation Research Board (TRB) Annual Meeting and the North Carolina A&T State 2002 Dwight David Eisenhower Historically Black Colleges and Universities (HBCU) Transportation Fellowship Panel Meeting.

Thanks are also extended to the many authors of position papers, volunteer reporters, and organizations that have supported, contributed and worked diligently on the activities leading to this report.

TABLE OF CONTENTS

This project was awarded in response to the research proposal solicitation offered through the North Carolina A&T State University Transportation Institute's Urban Transportation Institute (UTI).

The major focus of the research project was to model transit issues associated with hurricane evacuation planning. However, the specific objectives were to use a traffic operations based hurricane evacuation model to determine the expected time to evacuate the entire population; identify the locations of the potential traffic bottlenecks; and assess traffic operation strategies aimed at mitigating the resulting congestion. Since some sectors of the population, specifically the elderly and disabled, may not be able to evacuate by their own means, this project also presents a methodology to determine the scheduling of buses to be used for that purpose. The methodology is presented in the form of case studies that analyze the use of buses during a hurricane evacuation of a small urban and rural area located in the state of North Carolina.

There are many observations, conclusions and recommendations that surfaced from this research project. They include general comments to very specific suggestions aimed at enhancing travel speed and lowering travel time for specific evacuees within a small urban and rural area. In order to implement many of the items contained in this report, which we feel is needed in a comprehensive evacuation plan, several general observations and recommendations can be made.

First, mobility and accessibility of the under-represented population residing in both small urban and rural hurricane-prone areas will be enhanced. And second, institutional arrangements and inter-agency coordination between small urban and rural transit providers and other emergency responsive agencies located within the surrounding jurisdiction will also be improved. A lot can be gained in terms of minimizing loss of life during a natural disaster by providing alternative evacuation routing strategies prior to the actual disaster taking place. Interagency communication is critical to proving optimal evacuation success.

The need to evaluate multiple scenarios through simulated evacuation crises management environment cannot be over emphasized as it will provide initial training and knowledge base for the events that will unfold.

The results of the completed work consist of modeling transit issues unique to hurricane evacuations in small urban and rural areas located in the state of North Carolina. Particular emphasis is placed on identifying transit options needed to assist underrepresented populations such as the elderly and disabled, who maybe, in some ways, captive without transportation means or limited in terms of evacuation means. The value of this research is in its ability to provide North Carolina's small urban and rural emergency management agencies and officials with a practical methodology that facilitates local disaster planning and preparedness efforts regarding the movement of underrepresented citizens during evacuations.

In April 2000, the North Carolina A&T State University Transportation Institute's Urban Transportation Institute (UTI) funded this project as a means of investigating the modeling of transit issues that are unique to hurricane evacuations in small urban and rural areas located in the state of North Carolina. The Civil & Environmental Engineering Department in the College of Engineering at North Carolina A&T State University collaborated with consultants from the Electrical Engineering Department at Southern University and the Civil & Environmental Engineering Department at the University of Tennessee at Knoxville to perform the research.

This report constitutes the final report for the research project conducted by the Research Team for the North Carolina A&T State University Transportation Institute's UTI. This research project had several specific research objectives and they were:

• Use a traffic operations based hurricane evacuation model to:

• Determine the expected time to evacuate the entire population under a natural threat such as a hurricane;

• Identify the locations of the potential traffic bottlenecks; and

• Assess traffic operation strategies aimed at mitigating the resulting congestion.

• Present a methodology that uses the scheduling of buses for the purpose of transporting North Carolina's small urban and rural elderly and disabled citizens during a hurricane evacuation.

The investigation of these objectives and associated tasks are presented in Chapter 2.

The Research Team was comprised of Dr. Judy A. Perkins (Project Principal Investigator, North Carolina A&T State University), Dr. Ibibia K. Dabipi (Consultant, Southern University), and Dr. Lee D. Han (Consultant, University of Tennessee at Knoxville). Benjamin Hayes (Undergraduate Student, Civil Engineering), Melanie Hylton (Undergraduate Student, Transportation and Logistics), Tony Reames (Undergraduate Student, Civil Engineering), Jessica Rich (Undergraduate Student, Civil Engineering), Hector Williams (Undergraduate Student, Transportation and Logistics), and Rashad Wise (Undergraduate Student, Civil Engineering) are the other six key members of the research team.

One of the impediments identified during an evacuation operation occurring from a natural disaster, is the issue of transportation. The Federal Emergency Management Agency (FEMA) and several other federal, state, and local transportation entities agree that access to reliable transportation is a key factor in a person’s ability to move efficiently to locations of safety.

To address the safety concerns and the state of readiness of smaller urban and rural residents under the threat hurricanes, which do strike on a yearly basis, the proposed study sought to build upon the issue of transportation logistics during an evacuation. Particular emphasis was placed on modeling transit issues unique to an under-represented population residing in a small urban and rural hurricane prone municipality in North Carolina. Both the small urban and rural municipalities served as the bases for identifying needed transit resources and services as well as implementation steps for the successful delivery of these resources and services to an under-represented population, namely the elderly and disable residents.

To achieve the aforementioned research project, a number of tasks were identified for this study and the details of each are listed below.

In reviewing the current practice(s) surrounding the subject matter, the research team studied the details involved in planning for hurricane evacuation operations, the application of modeling technologies used in the planning process, and the current transit practice (s) associated with evacuation operations in the State of North Carolina. These practices were compiled and compared to other coastal states as well as the recommended approaches discussed in the literature.

The research team gathered both quantitative data and qualitative information on existing demographic, infrastructure and geographic data to name few. This data as it relates to hurricane evacuation operations was compiled and organized for use in Task 5. While formal interviews were not solicited nor conducted, local anecdotes from citizens and officials were not discouraged either.

To facilitate the selection of sites, a set of criterion were developed in the context of addressing human and transit resources and services used as well as needed during an evacuation operation. Elements of the criteria included the identification of characteristics that define small urban and rural municipalities; existing dispersion vs. concentration of transit provider net- work; and existing dispersion vs. concentration of the under-represented population to name a few.

Two hurricane-prone areas were selected, one small urban and one rural, as case studies and for gaining a more in-depth understanding of the residents and transit needs during a hurricane evacuation as well as for modeling purposes. It was important for the selected sites to be reasonably representative of other small urban and rural areas of the State in order to get good impression of the magnitude of the problem. This process determined how the sites were selected.

A traffic operations based hurricane evacuation model was employed to analyze the selected sites by simulating a variety of evacuation scenario activities based on quantitative data collected in Task 2. Particularly, the use of existing transit facilities and public fleets was considered. In addition, a combination of readily available technologies such as global position systems (GPS) and dynamic vehicle routing and management were investigated. Other tools such as Geographic Information Systems (GIS) were also studied for flood extent estimations. The results from the various evacuation scenarios were then assessed.

Figure 2.1 - Hurricane Evacuation Model Screen Display

Evaluation mechanisms were established to assess the validity of the model results and integration of new technologies. Several necessary elements were evaluated and the following were suggested: identifying future needs of the selected sites in terms of evacuation planning and operations; protocol structure of a demand responsive service; transit provider performances; cost-effectiveness of evacuation plans and alternatives; and accessibility to the under-represented population to name a few.

An assessment of intelligent transportation system (ITS) technologies that can benefit hurricane evacuation operations was conducted. The research team identified ITS technologies that could augment initiatives deployed by North Carolina’s Department of Transportation. Recommendations were made on how to integrate these advanced technologies into the results from Task 5, thus improving the mobility and accessibility of elderly and disable citizens during a time of an evacuation.

The results from the previous tasks are detailed in the following chapters, and will be submitted to North Carolina A&T State University's Transportation Institute, and may be submitted for publication and/or conference presentation. The report suggests a methodology for addressing the transit issues unique to hurricane evacuation of elderly and disable residents of small urban and rural areas. If deemed beneficial, the findings may be shared through technology transfer activities such as continuing education and training courses suggested herein.

According to FEMA, during the 20th Century more than 10,000 people lost their lives to hurricanes in the US alone. The fatality rate has been dropping thanks to early warnings and proactive evacuation activities. However, lives are still threatened and sometimes lost when natural disasters strike. While larger metropolitan areas often have alternative evacuation plans developed in detail, smaller urban and rural municipalities often have less than comprehensive plans for evacuation operations or emergency responses. This research modeled transit issues unique to hurricane evacuation of elderly and disable citizens residing in both a small urban and rural municipality in the state of North Carolina.

During a hurricane evacuation a large number of vehicles have to be moved across a road network in a short period of time [33]. The number of vehicles leaving becomes a big problem for areas such as Florida, Georgia, Louisiana, South Carolina, North Carolina, Texas, and Virginia to name a few. The number of evacuating vehicles varies depending on the number of residents, intensity of the hurricane, location, time of occurrence, direction of approach, and the number of tourists' [33].

Moreover, the evacuation of hurricane-vulnerable people is largely a transportation problem. While it is possible to have several days warning prior to the hurricane arrival, as the arrival of the hurricane can be to some extent predicted, it is important to begin preparations for the hurricane well in advance of any information that it may land in a location near you. To this extent, the critical factors that determine a community's response to a disaster are the preparedness measures, such as emergency procedures, evacuation plans, search and rescue training, and effective telecommunication systems [5]. Therefore, the thrust of this chapter was to provide a review of pertinent theoretical and empirical literature that investigates both the planning of hurricane evacuation operations as well as the technological applications used in modeling hurricane evacuation operations. Accompanying the literature review, the research team inventoried a number of initiatives and evacuation plans of communities, agencies and organizations addressing the issue of mobility and accessibility as well as the use of transit resources as a means to evacuate the elderly and disable citizens residing in hurricane- prone areas.

According to the New Orleans Citizens Emergency Preparedness Guide, a hurricane is a cyclone (low-pressure system) developing in the tropics with a minimum wind speed of 74 miles per hour. The wind rotates in a counterclockwise direction around the center of the storm, called the "eye", where the winds are nearly calm. The wind in an intense hurricane may exceed 150 miles per hour with gusts above 20 miles per hour just outside of the storm's center. Hurricane force winds may extend out 100 miles from the center with gale force winds (39 miles per hour or higher) extending outward 250 miles. Bands of very intense thunderstorms spiral outward from the eye of the hurricane for several hundred miles producing torrential rain, and occasionally spawning tornadoes when they begin moving over land.

Moreover, a hurricane can last for two weeks or more over open water and can run a path across the entire length of the Eastern Seaboard. When a hurricane watch is issued, the best response is to protect your property by boarding up windows, bringing in outside items, and being prepared to evacuate the areas as soon as officials advice. Evacuation ensures that virtually nobody is present.

Table 3.1 - Saffir-Simpson Hurricane Scale

                                    Source [3]

Hurricane categories as per Table 3.1 are characterized by its intensity which is embodied in the Saffir-Simpson scale: Category 1 (Minimal Damage) - no real damage by wind is caused to buildings. Some damage is done to poorly constructed signs. Some damage primarily occurs to unanchored mobile homes, shrubbery, trees, and foliage. Low-lying roads are inundated by storm surge. Minor pier damage occurs; Category 2 (Moderate Damage) - roofing, doors and windows of homes and businesses are damaged by winds. Considerable wind damage is done to mobile homes and vegetation. Low-lying roads are inundated by storm surge. Considerable damage is done by storm surge and wave action to piers. Small craft in unprotected anchorage break their moorings; Category 3 (Extensive Damage) - winds cause structural damage to homes and utility buildings with a minor amount of curtain wall failure. Mobile homes are destroyed. Storm surge flooding destroys many smaller buildings while floating debris damages large buildings. Terrain continuously lower than 10 feet above mean sea level will flood; Category 4 (Extreme Damage) - more extensive curtain wall failures with some complete roof failure on homes occur. Major damage is caused to lower floors of homes and businesses from storm surge flooding. Terrain continuously lower than 15 feet above mean sea level will flood; and Category 5 (Catastrophic Damage) - there is complete roof failure of many homes and business as well as complete building failure of small structures. Buildings located less than 20 feet above mean sea level will experience major storm surge flooding to lower floors [3].

According to Table 3.2 between 1989 and 2000, there were 19 different hurricanes that struck the continental United States. During this period, approximately 63 percent were Category 1 and 2 hurricanes, 37 percent were Category 3, 4, and 5 (major hurricanes), and 47 percent made landfall in various parts of North Carolina. The 1995, 1996, 1998, and 1999 seasons were the most active in that hurricane activity grew to a higher level, thus producing many of the most damaging major hurricanes on an annual basis.

Table 3.2 - Hurricane Tracking for US and North Carolina

Source [4]. NOTE: *Not all of the States listed were directly affected by the hurricane. Some were affected either by tropical storms, tropical depressions, or subtropical storms prior to or after the hurricane. The path and strength of the hurricane plays a major role in this.

Moreover, Florida is the most hurricane prone state in the US with the southeast being the most at risk. Eleven hurricanes made landfall in Florida, four in Georgia, five each in Louisiana and Virginia, six in South Carolina, and nine in North Carolina. The costliest hurricane recorded was Hurricane Andrew in August 1992. It hit the States of Florida, Georgia, and Louisiana with damage estimates topping $25 billion dollars. During the 1996 season, Hurricanes Bertha and Fran both hit the state of North Carolina. Total costs associated with damages for the US was approximately $3.5 billion with $1.4 billion allocated to the state of North Carolina. The number of deaths directly attributed to hurricanes occurring from 1995 to 1999 inclusive for the United States totaled 135 death verses 64 deaths in North Carolina.

Figure 3.1 - Hurricane Fran

While hurricanes are inevitable on the East Coast of North America, developing intelligent responses in the years, months, days, hours, and even minutes before a hurricane landfall can enable individuals to prepare effectively, thus limiting human and material losses.

As per the leading authors in evacuation planning and emergency management concepts, planning hurricane evacuation planning operations are often managed at the local levels and have historically fallen under the jurisdiction of emergency planning and law-enforcement agencies [25]. Additionally, many of the entities gave little attention to planning in a manner that facilitates utilizing the transportation system to its fullest potential when the need arises. However, there are some entities that have used a comprehensive and/or cooperative approach to evacuating local citizens in the event of a hurricane. This section of the chapter will highlight these practices.

In 1987, the Department of Energy contracted with Oak Ridge National Laboratory to assess issues and criticisms of evacuation planning for all hazards under an integrated emergency management concept. The work identifies gaps in knowledge about evacuation planning issues and research that can address these gaps. Over 300 documents were reviewed and abstracted, and key findings were summarized. Issues were identified by review of hearings, litigations, critiques, and discussions with planners and experts. Evacuation planning for some hazards has integrated physical, risk studies with quantitative evacuation traffic modeling and behavioral research to produce comprehensive planning guidance. The adoption of an integrated or generic emergency management approach has bolstered and will further bolster the expediency of evacuation planning. Over the past ten years, most aspects of evacuation logistics have been identified and researched. There are indications that the local implementation of evacuation procedures has improved. People who could have evacuated to safety continue to die in disasters. The research identified ten major issues in evacuation planning that cut across hazards. Planning for large-scale evacuations requires improvement. A better understanding of special evacuation planning needs for fast-moving events is needed. Evacuation planning for concurrent hazardous events is lacking. A better understanding of human behavior in evacuations is desirable. The accuracy of evacuation time estimates should be established. Guidelines on reentry after an evacuation should be improved. Special populations planning require further investigation. Liability for evacuation decisions should be resolved. Uncertainties and problems in evacuation decision-making need greater attention. Adoption and implementation of integrated evacuation plans need further investigation.

McLeod observed a number of many problems (e.g., congestion, high winds, downed power lines, and sometimes slowdowns due to the collection of tolls on an evacuation route that surfaced with the Florida's transportation system during hurricane evacuations. More specifically, the three major elements contributing to Florida's transportation problem when evacuating hurricane-vulnerable people are (a) physical capacity (e.g., number of lanes), (b) structural (e.g., drainage and washouts), and (c) traffic management (e.g., signals and managing accidents). In an effort to resolve this problem, a number of recommendations were made. They were: (1) no major roadway improvements (widening or new road) should be built strictly due to hurricane evacuation needs, (2) the concept of armoring, protecting, and moving coastal highways for evacuation purposes should be considered only on a case-by-case basis. The random event of a hurricane is such that it is usually cheaper to rebuild roads as needed, rather than move or rebuild large portions of our coastal highway system, (3) Florida Department of Transportation (FDOT) should not fund construction or planning for any facility that intrudes into a Federal Coastal Barrier Resource System, (4) FDOT should eliminate toll collection during a hurricane emergency. Specific language should be added to all new bonding terms of indenture so that suspension of tolls is legal and understood, and (5) emergency management officials must couple the hazards data provided by the National Hurricane Center with clearance time information calculated from transportation analyses. By considering these sources of timing data, officials can determine when a strong evacuation advisory or order must be issued to allow people time to reach safe shelter.

During September 1999, Hurricane Floyd skirted the Southeast Coast of the United States and made landfall in eastern North Carolina. During the storm's life, an estimated 3 million people evacuated from their homes in the states of Florida, Georgia, South Carolina and North Carolina. This storm occurred two weeks after Hurricane Dennis made landfall in eastern North Carolina. Hurricane Dennis saturated the ground before Hurricane Floyd struck. This resulted in unprecedented flooding in that part of the state. Hurricane Floyd was a Category 4 storm that closely followed the path of Hurricane Hugo in 1989, until 2 days before it made landfall. For these reasons and many other reasons, the affected population took this storm more seriously than past storms [29].

While there were numerous public complaints about the evacuation process, the evacuations in all states accomplished their primary purposes: (1) evacuees from the vulnerable areas of each state evacuated before the damaging weather arrived; and (2) the evacuation occurred without loss of life. Moreover, the impact of Floyd on the transportation system offered officials a window to identify some areas of improvement [29].

In October 1999, the US Army Corps of Engineers contracted PBS&J to conduct studies to determine the nature and causes of the transportation problems encountered during an evacuation. Also, they were charged with developing specific recommendations and tools to assist the emergency management community in the states of Florida, Georgia, South Carolina and North Carolina to reduce or eliminate transportation problems in the future [29].

The findings from the study identified two specific tools that would assist transportation officials in the management of the hurricane evacuation planning process. They were: (1) use strategies to transpose "coast-bound" lanes of travel to the direction of evacuating traffic on controlled access facilities, such as Interstates highways, and (2) use ITS applications that will better manage existing roadway capacity, manage traffic congestion and provide more information to the affected public. Complementing the ITS recommendation, there were a number of non-ITS technology applications that were endorsed, such as availability of fuel, restaurants, rest areas, and welcome centers remaining open during the critical evacuation period [29].

Emergency evacuation is one of three major points Florida hopes to address with their Mobility 2000 Initiative. The other two are expansion of trade and tourism quarters and relieving traffic congestion. Moreover, Dick Kane, FDOT spokesman, states that Hurricane Floyd caused the largest evacuation in US history (approximately 2 million people). With the passage of Mobility 2000, the largest transportation funding package in the history of Florida, the emphasis will not just be on building more roads, but on doing a better job in communicating with the public, providing better signage and more advance notice. A few of the projects advanced under the Mobility 2000 Initiative specific to emergency evacuation were: (1) four- lane sections of route 87 and 79 in northwest Florida, (2) six-lane sections of Interstate 95 from Daytona Beach to the Georgia state line, and (3) six-lane sections of Interstate 75 in Lee and Collier counties. These projects, in addition to others, are part of Florida's progressive $6 billion dollar transportation improvement investment that Governor Jeb Bush strongly endorses because the transportation system is vital to the economic prosperity and quality of life of Florida's residents' [19].

Regarding the City of New Orleans, Wolshon describes the city of New Orleans' hurricane evacuation plan. Specifically, he highlights the problems that are particular to implementing a regional evacuation plan for the city as well as the creative ways that disaster and transportation planning agencies can work together to solve them. Also, several critical operational, strategic, and administrative issues affecting the movement of New Orleans citizens during the 1998 Hurricane Georges evacuation were addressed. Some of the assumptions and findings include: (1) implementing the use contraflow evacuation technique to reverse traffic flow on freeway routes to move all traffic in an outbound direction; and (2) recognizing that while a total evacuation is required, an evacuation of only part of the population is practically possible. In fact, of the 1.4 million citizens residing in New Orleans, it is assumed that 60 percent of the population will want, or be able to leave the city. The primary reasons are a lack of access to transportation (it is estimated that about 200,000 to 300,000 people do not have access to reliable personal transportation), an unwillingness to leave homes and property (estimated to be at least 100,000 people) and a lack of outbound roadway capacity.

Figure 3.2 - Contraflow Traffic Movement

To complement Wolshon's research, the City of New Orleans Office of Emergency Preparedness developed the Citizen Emergency Preparedness Guide. The purpose of the guide is to provide citizens with information on: (1) the types of hazards that confront the city of New Orleans, (2) preparing their own individual emergency plan, and (3) the proper procedures that should be followed in the event of an emergency. Also, the Orleans Parish three-phase evacuation plan is discussed in detail. The phases are precautionary, recommended, and mandatory. During the precautionary phase, the focus is on people who are most vulnerable to hurricane and the effects of both water and wind. These are offshore workers, persons on coastal islands or in wetlands, person aboard boats, and those living in mobile homes and recreational vehicles. No special traffic control or transportation measures are implemented, however, procedurally the following occurs: (1) a state of emergency is declared, (2) the City of New Orleans Emergency Operating Center (EOC) is activated, (3) special facilities, including nursing homes, begin preparation for possible evacuation, (4) staging areas and/or shelters are announced, (5) citizens implement their personal evacuation plan at their discretion, and (6) persons with special needs are encouraged to call on Social Service Agency, family and friends to help them as well as make them part of their plan. During the recommended phase, it has been determined that the storm has a high probability of causing a significant threat to people living in the areas of risk. Government authorities will recommend that persons at risk evacuate. Staging areas are designated for persons needing transportation. Specific procedures during this phase are: (1) the City of New Orleans EOC is staffed for 24-hour operation, (2) local transportation will be mobilized to assist persons who lack transportation, (3) bus routes and locations of staging areas for those needing transportation to shelters in or out o the Parish, will be announced via radio and television, and (4) relative and neighbors should help family and friends who need transportation and other assistance. Lastly, during the mandatory phase, authorities put maximum emphasis on encouraging evacuation and limiting entry into the risk area. This is the most serious phase of evacuation. The State Office of Emergency Preparedness, State Police, State Department of Transportation and the Louisiana National Guard will assume coordination and responsibility for traffic control on all major evacuation routes. Because of deteriorating weather conditions, at some point, evacuation routes will be closed and the remaining people at risk will be directed to a last resort refuge. Procedures carried out during the mandatory phase are: (1) persons living in designated evacuation zones will be instructed to leave, (2) traffic controls will be imposed to direct persons to designated evacuation routes; and (4) emergency alerting system (EAS) radio stations 870 AM, 101.9 FM and the news media will issue evacuation information.

The Harris County Emergency Management Office in Texas developed their evacuation plan to include specific guidelines to move people without transportation and people who require assistance (e.g., blind, disable, elderly, and non-English speaking residents). Also, some attention was given to the evacuation of special facilitates (e.g., hospitals, jails, institutions for the handicapped or disable, and nursing homes). Finally, there were four accessibility and mobility strategies suggested. They are: (1) use public transportation to pickup citizens requiring assistance at their residence in case of an evacuation, (2) use the pre-established Metro bus stops on local Metro bus routes as pickup points for able-bodied residents without transportation, (3) identify transportation pickup points for the rural areas of Harris County not serviced by Metro and announce through the Public Information Officer, and (4) require the special facilities to have their own evacuation plans outlining the transportation resources needed to conduct the evacuation, as well as the requirements for other resources such as food, medicine, and attendants. The officials operating these facilities are encouraged to make arrangements with suitable reception area facilities to lodge their clients.

In the state of North Carolina, the North Carolina Emergency Management Division (NCEMD) develops the evacuation plans and the North Carolina Department of Transportation (NCDOT) has a secondary role in the effort. Issuing evacuation orders are local county decisions. The local NCDOT managers keep in touch with local emergency management officials at all times during the evacuation operations.

Moreover, as per the NCEMD and Table 3.3, currently there are 49 North Carolina counties that provide hurricane safety and preparation tips via their respective emergency management/service agency's websites. The level of information associated with each website varies across counties and the details regarding the resources allocated to transporting elderly and disabled individuals is extremely limited. However, some counties as per their responses to an email questionnaire did cite some of their progressive activities and they were; (1) Cabarrus County contracts with the Laidlaw Transportation, Incorporated van system. In time of an evacuation, these vans along with school buses would become part of the County's Emergency Operations Center transportation assets and made available for use as needed. Ten of the 24 multi-passenger vans are lift-equipped to assist with servicing the special needs of individuals; (2) Mecklenburg County's emergency operations center is set up by the City of Charlotte Emergency Management agency. The Charlotte Area Transit System (CATS) provides transportation and other evacuation services as needed during emergency situations. Currently, CATS is on the back-up support list for the police, fire, medic, and other emergency service agencies. CATS Bus Operations Division maintains total control over the bus fleet while offering support as requested. Moreover, CATS do not have a published plan outlining their support activities. Their actions would be responding to servicing request by the emergency operations center and other emergency service entities; and (3) Guilford County incorporates both the Greensboro Transit Authority (GTA) and other internal county transit systems into their citywide evacuation plan. Additionally, transporting the elderly and disabled is supplemented by the National Guard when permitted by the NCEMD and the County's Special Needs Team (headed by Social Services) is responsible for transporting patients, clients, records and medicine from facilities under their jurisdiction.

Table 3.3 - NCEMD Local & County Emergency Management Entities

                    Source [24]

Several analytical tools have been used to facilitate major issues in hurricane evacuation planning for a variety of natural disasters. During the 1990 - 2000 time frame, FEMA spent approximately $20 billion dollars to help people repair and rebuild their communities after the occurrence of several natural disasters. In response, FEMA took a proactive approach to change the way the public deals with disasters. In October 1997, FEMA launched a nationwide initiative entitled, "Project Impact: Building Disaster Resistant Communities." The initiative was designed to help communities protect themselves from the devastating effects of natural disasters with preventive actions that dramatically reduce disruption and loss [27]. More specifically, the initiative commonsense approach was based on three principles--preventive actions must be decided at the local level, private sector participation is vital and long-term preventive efforts and investments are essential.

Currently, 120 communities throughout the United States are participating in Project Impact. FEMA is collaborating with the Environmental Systems Research Institute (ESRI) to foster and support the use of GIS technology as a tool for implementing disaster resistance, particularly as it pertains to developing or implementing strategies for reducing damage from natural hazards. Additionally, Project Impact is serving to empower communities as they pursue smart growth strategies while strengthening the federal government's partnership with local government [27].

Regional evacuation is often a viable safeguard against natural catastrophes such as hurricanes and volcano eruptions, and man-made disasters such as toxic gas release into the atmosphere. Several computer programs exist for modeling regional evacuation on a given road network. As would be expected, these models are data intensive. In 1992, a personal computer (PC)-based interactive model, Software Utility for Regional Evacuation (SURE), to create input to Interactive Dynamic Evacuation Model (IDYNEV) and analyze its output was developed. The preprocessor module of SURE substantially reduces the time required to create input files to IDYNEV and the post-processor module summarizes the simulated results from the usually huge IDYNEV output files. The model also illustrates the evacuation network graphically and highlights problem links and intersections at selected intervals. SURE is being used extensively for evacuation studies under the US Army's Chemical Stockpile Emergency Preparedness (CSEP) Program.

In FY 1995, after Hurricane Iniki caused $260 million in damage to Hawaii, Congress authorized the funding for a Pacific Disaster Center (PDC). The PDC supports federal, regional, and international emergency managers along the Pacific Rim by providing high quality, timely imagery, maps and other value-added information products about natural and human induced disasters, including hurricanes, tidal waves, floods, and earthquakes via the Internet.

Results to a 1998 project performed for the FDOT by the Center for Urban Transportation Research (CUTR) at the University of South Florida were presented in a report separated into two sections comprising the two phases of the project: Phase I - Hurricane Evacuation Traffic Analysis; (Analysis of Traffic Demands During Historical Evacuations); Phase II - Advanced Technologies for Improving Operations Management During Hurricane Evacuations (Traffic Surveillance Camera Resources Summary; Examination of Real Time Traffic Information on World Wide Web (WWW) Sites. This report presents the findings of all three elements of the project.

In June 1999, the South Carolina Department of Transportation (SCDOT) started developing a hurricane management system to manage evacuation and track road usage in a natural disaster. With the help of Intergraph, the agency launched the Hurricane Evacuation Decision Support System just prior to Hurricanes Dennis and Floyd. The system is a Web-based presentation of maps that incorporates live information from SCDOT's GIS, remote traffic counters, evacuation routes, detour maps, and real-time weather data. It is designed to put rapidly changing traffic and weather information at the fingertips of state officials who manage evacuations. Additionally, SCDOT personnel customized the charts for Floyd to isolate traffic on specific routes, such as Interstate 26 heading west across the state, or in metropolitan areas, such as Charleston, where many evacuation routes merge toward inland destinations.

Referencing the June 2000 PBS&J study mentioned earlier, a web based evacuation travel demand model was constructed so that major traffic congestion areas and traffic flows could be anticipated and monitored for a Floyd type event. Microsoft's Visual Basic and ESRI's Map Objects, Map Objects IMS, and ArcView developed this important tool. Out of county evacuation data calculated in FEMA, the US Army Corps of Engineers hurricane evacuation study products and Regional Planning Council hurricane studies were used as a baseline for the model. The model was set up so that a state could dialup the model and input (hurricane category, expected participation rate/compliance rate, tourist occupancy, and destination percentages) information pertaining to their area. Once the information is submitted, the model updates a number of system forecasts and graphics and provides key information for a significant hurricane threat. Other data that can be input real time are hour by hour traffic counts at locations where a state or county have the ability to collect and communicate such information. The major model outputs are: (1) expected congestion levels by highway segment, (2) tables of expected vehicles crossing state lines by direction, (3) comparisons of traffic count station data to forecast condition, (4) numbers of vehicles generated by each county traveling to specific inland locations, and (5) route information by segment including a number of lanes, facility type, service volume, and congestion measure [32].

In March 2000, PDC contracted with ESRI to implement dynamic mapping. This technology will enable members of the fire department or other local emergency response teams to indicate the location of an event by clicking a point on the map, thus increasing their exposure to more information and enhancing their emergency management activities. This is an upgrade in technical capabilities since the previous maps and images on the PDC site were static in nature. Finally, the PDC's future plan endorses the use of interactive mapping and satellite images. This feature enables an array of users to facilitate the sharing and use of a vast amount of information via the Global Disaster Information Network (GDIN) [20].

Currently, North Carolina uses an interactive web based system called the Traveler Information System (TIMS) (see Figure 3.3 [38]) to provide the viewer with a statewide map that is divided into three regions-Coast, Mountain, and Piedmont. Clicking on either of these regions will link to county maps within the region. Further clicking on the county map reveals the county road network and corresponding detailed road closure information text. Information accuracy and updates of road closure due to incidents and weather conditions remains the responsibility of public or private organizations providing it.

Figure 3.3 - Traveler Information System (TIMS)

Typically when people think of public transportation they think of big buses in big cities, however public transportation comes in many forms (bicycles, buses, carpooling, trains, vanpooling, and vans) and in many areas--rural and urban [13]. For the purpose of this research, our discussion on public transportation will be limited to the use of buses and vans.

Buses come in many varieties. They range in size from 24 feet in length to as much as 60 feet in length. The size of the vehicle is dependent upon several factors: (1) the maximum load that may be on a bus at one time, (2) the width of the streets in neighborhoods, and (3) sometimes mandates from governing boards about bus sizes. The typical bus lengths are 35 and 40 foot buses that transport anywhere from 37 to 64 passengers. Depending upon the size of the community being served the sizes may be smaller or larger [13]. The various transit properties will usually run buses on a regular fixed route with fixed schedules and stops. Buses may run anywhere from every 10 minutes to every 60 minutes depending upon the time of day and the community being served.

In compliance with the Americans with Disabilities Act of 1990 all new buses are purchased with either wheelchair lifts or ramps. The lifts are located in either the front or rear door. The lift stows away underneath the steps so those passengers who don't need the lift can enter the bus. In the case of a bus with a ramp, there are no steps to go up. The floor is flat and this makes entering the bus easier for people with disabilities, the elderly, and people traveling with small children [13].

Vans are often used in rural communities to provide transportation. The types of vans used are standard 15-passenger vans, 7-passenger mini-vans, and conversion vans. A conversion van is a passenger van that has been altered to allow the addition of a wheelchair lift and securement stations within the vehicle. Generally these vans have higher roofs and are sometimes referred to as "turtle tops" [13].

The vans generally provide a combination of rural general public and/or human services transportation. Some systems have developed "routes" that they travel at certain times and on certain days of the week to provide public service. Most systems provide human services transportation only. This means that the users of the service must be clients of an agency that contracts with the system [13].

The fastest growing segment of the population nationally and within the state of North Carolina is the elderly. In fact, North Carolina's elderly population is growing at an even faster rate than the nation as a whole. North Carolina's population also includes many citizens who are disabled. Many of these citizens are able to use public transit services for health and social reasons as well as to assist them in getting to and from work. Also, for those that are able to use public transportation services, this may be their only way to get to a doctor's office, grocery store, or church [41].

Clearly, mobility options for the elderly and disabled citizens are extremely important and can become problematic if public transportation did not exist. Public transportation do, in fact, provide more than just an option to elderly citizens, people with disabilities and others without access to personal vehicles or those who are unable to use them--it is their vital link to the community [37]. Public transportation provides crucial access to services and can, in the long run, reduce public costs for the delivery of those services. Lastly, public transportation is strategically positioned to elevate its role in providing mobility and access to a safer environment during a hurricane evacuation, particularly for individuals with special needs.

Implication of Findings

Indeed, planning for hurricane evacuation operations is a process that will continuously evolve. While the nuclear power facilities emergency preparedness planning can serve as an example for planning hurricane evacuation operations, it is important to develop progressive comprehensive plans that articulate responses well before a hurricane make landfall. This is key to facilitating entities with emergency-related responsibilities to coordinate transportation resources more efficiently as well as enabling individuals to prepare effectively. The comprehensive plans should explicitly address the issue of mobility and accessibility in a manner that integrates activities across several jurisdictions, thus raising the level of understanding regarding uncertainties and problems involved in evacuation decision-making. Finally, planning for special populations to be evacuated is paramount. It has been stated that evacuation ensures that virtually nobody is present. In order for this statement to hold true, proactive initiatives complemented with technology must carefully address the lack of mobility and accessibility that confronts a very large percentage of our citizens.

By combining results from the literature review and the canvass of public and private hurricane evacuation operations initiatives, a set of factors that could greatly influence the selection of sites to Beta test the hurricane evacuation model were developed.

The State of North Carolina has a total area of 52,712 square miles and is divided into four physiographic provinces (Tidewater Coastal Plain, Upper and Middle Coastal Plain (UMCP), Piedmont, and the Mountains). The Coastal Plain land and water areas comprise nearly half the area of the State. The Tidewater portion is generally flat and swampy while the interior is gently sloping and, for the most part, well drained. The Piedmont portion rises from the 200-foot level to near 1500 feet at the base of the mountains. The land is mostly gently rolling hills with a great deal of hard rock near the surface. Also, the Piedmont comprises about one third of the State. Lastly, the Mountains comprise the smallest area covering a little more than one fifth of the State [24].

Figure 4.1 - State of North Carolina Physiographic Provinces

North Carolina enjoys a mild but varied climate owning to its wide range in elevations and to inland areas' distance from the sea. The average temperature varies more than 20 degrees from the lower coast to the highest mountains. The average annual rainfall of 80 inches in the southwestern mountains is the greatest in the eastern United States. East of the mountains, average annual rainfall is 40 to 55 inches. Winter storms, generally approaching from the southwest, are triple threats for North Carolina. Depending on their exact tracks, they deliver heavy snow, severe icing or heavy rain and sometimes cause flooding. The entire state is susceptible to frontal and air mass thunderstorms. On average, North Carolina experiences 14 tornadoes per year. These storms might occur at any time, but are most likely in the spring and summer months. During each hurricane season (June - November), the North Carolina coast likely will be threatened, if not struck, by at least one tropical storm. For climate purposes, North Carolina is divided into 8 regions and they are: (1) Southern Mountains; (2) Northern Mountains; (3) Northern Piedmont; (4) Central Piedmont; (5) Southern Piedmont; (6) Southern Coastal Plain; (7) Central Coastal Plain; and (8) Northern Coastal Plain [24].

North Carolina's overall population is estimated at over 7.7 million spreads over a total land area of slightly more than 31 million acres. Approximately one third of the State's total population lives in the Coastal Plain; however; the population is not evenly distributed throughout the area. Most of the Coastal Plain population is concentrated in the southern section where the major military installations are located. Even though the Piedmont physiographic province is one-third the State's total area, over one half the State's population resides in this area. The population concentrated in the Piedmont Crescent, a large, loosely defined area that stretches from Charlotte through the Statesville, Winston-Salem, Greensboro-High Point area, and Burlington to the Raleigh-Durham-Chapel Hill area. The Mountain region is the smallest and least populated section in the State. About fifteen percent of the State's population lives in the mountain region [24].

Moreover, residents age 65 or older represent approximately 12.1 percent of the State's total population. Residents that are disable, living in households without personal vehicles, or are beneficiaries of social security represents approximately 6.8 percent, 9.6 percent, and 20 percent of North Carolina's population, respectively.

Vital to North Carolina's economy is their transportation network. The network consists of airports (public and private), ferries, ports, public transportation (buses, vanpools, etc.) operating in rural and urban areas, rail (passenger and freight), roads (Interstate, State routes and US routes), and a few miles of navigable waterway corridors. While each element of the State's transportation infrastructure network is important, the road system and public transportation system are more heavily relied upon during a hurricane evacuation then any of the other travel modes. Within North Carolina, there is approximately 78,000 miles of roadways. The primary highways total mileage is 14,616 and it consist of 12 Interstate Highways, 226 State routes, and 74 US route. Total mileage for the secondary roads is 63,467. With regard to public transportation systems, there are more than 100 urban and rural properties operating throughout the state. There are 18 public-owned, urban transit systems and about four million North Carolinians have access to rural public transportation services operating throughout the 100 counties. Clearly, the need to maintain adequate, well-maintained roadways, bridges along all routes, and capitalize on the use of public transportation is key to not only providing direct access to emergency shelter locations, but also facilitating an evacuation response to severe weather and other civil emergencies.

Figure 4.2 - North Carolina Public-Owned Public Transportation Systems

In the wake of Hurricane Floyd, 66 counties in North Carolina were declared disaster area by FEMA. Over 227 shelters in addition to motels, etc. were used by more than 100,000 people.

Figure 4.3 - Counties Declared Disaster Areas

The report identified some potential evacuation sites that may benefit from a coordinated evacuation planning effort that takes advantage of transit vehicles. Some of the criteria for site selection include:

1. Small urban and rural areas;
2. At risk of flooding and/or other hurricane threats;
3. Located entirely in the Tidewater or Piedmont areas (see below);
4. Captive population, i.e., low income, elderly, handicapped, etc.;
5. With transit service somewhere nearby;
6. Availability of demographic information; and
7. Evacuation was deemed necessary during Floyd.

Based on the guidelines presented, a number of potential sites were identified by:

• Literature review through hundreds of websites, news articles, government reports, and anecdotal stories to identify townships, areas, counties in North Carolina with records of flooding and other difficulties during Floyd.

• Identification of the locations of the at-risk areas to insure the selection of those in the Tidewater or Piedmont areas.

• Identification of locations not identified specifically in news but otherwise fitting to the criteria.

• Compilation of a list of 10-candidate analysis sites.

The results of the effort and solicited input and suggestions formed the basis for the final selection of the two eventual project sites.

After painstaking search of flooding areas due to hurricanes, more than fifty locales were identified and tabulated in alphabetical order in the Table 4.2. Unfortunately, most of these locations fall in the MUCP region. Only four areas were eligible, based on the selection criteria, to be considered as candidate sites. This prompted the question of whether Criterion 3: Located entirely in the Tidewater or Piedmont areas, which was identified on the previous page, was needed. The following ten sites will attempt to reflect the original criteria. It will be interesting to study the entire state in a follow up study in which Criterion 3 is eliminated and a broader definition for areas of interest is used.

Table 4.1 - Initial Compilation of Candidate Sites

Table 4.2 - Expanded Compilation of Candidate Sites