Risk perception and evacuation decisions of Florida tourists under hurricane threats: a stated preference analysis

Though most hurricane evacuation studies have focused on residents, tourists are also a vulnerable population. To assess their perceptions of risk and evacuation likelihood under different hurricane conditions, we surveyed 448 tourists visiting central Florida. Respondents viewed four maps emulating track forecast cones produced by the National Hurricane Center and text information featuring variations of storm intensity, coast of landfall, centerline position relative to the survey site, time until landfall, and event duration. We performed chi-square tests to determine which hurricane conditions, and aspects of tourists such as their demographics and previous hurricane experience, most likely influenced their ratings of risk and evacuation likelihood for respondents located on Pinellas County beaches or inland near Orlando, FL. Highly rated scenarios featured a Category 4 hurricane making landfall along the Gulf Coast with the centerline passing over the sampling site. Overall, tourists that indicated the highest risk and evacuation ratings were not previously affected by a hurricane, had a trip duration of less than 6 days, and had checked for the possibility of a hurricane strike before departure. However, results for other tourist attributes differed between tourists in coastal and inland locations. We found that although somewhat knowledgeable about hurricanes, tourists misinterpreted the track forecast cone and hurricane conditions, which led to a lower perception of risk and subsequent likelihood to evacuate. Tourists, particularly those from outside of Florida, need to be better educated about the risks they face from hurricanes that make landfall.     

The disruption and recovery of fish communities in the Indian River Lagoon, Florida, following two hurricanes in 2004

During the summer of 2004, four hurricanes (Charley, Frances, Ivan, and Jeanne) affected Florida between August 13 and September 27. Two storms (Frances: category 2 and Jeanne: category 3) made landfall in the southern portion of the Indian River Lagoon (IRL) on the east-central coast of Florida. The presence of Florida Fish and Wildlife Conservation Commission's long-term fisheries monitoring program in the IRL provided a unique opportunity to examine the effects of large tropical events on estuarine fish communities. Increased sampling efforts to monitor the effects of tropical disturbances on the fish community within the IRL and one of its major tributaries (St. Sebastian River) were initiated within days after the passing of the last hurricane (Jeanne). The objectives of the study were to characterize changes to the composition of the fish community within the lagoon and river immediately after the passage of two hurricanes, and to examine the recovery of the fish communities. Analyses indicated that immediately after the last hurricane passed, community diversity within the estuary decreased following these storms due to the absence of many marine species, whereas the fish community within the St. Sebastian River shifted to one containing a greater percentage of freshwater species. Recovery of the community structure to pre-hurricane conditions was evident within several weeks following the last hurricane, and by mid December 2004 (ca. 3 mo after the last storm), there was little difference between the pre-hurricane and post-hurricane fish communities.     

A 1200-year proxy record of hurricanes and fires from the Gulf of Mexico coast: Testing the hypothesis of hurricane-fire interactions

We present here the first high-resolution pollen record of vegetation response to interactions of hurricane and fire disturbances over the past 1200 yr from a small lake in Alabama on the Gulf of Mexico coast. The paleotempestological record inferred from the overwash sand layers suggests that the Alabama coast was directly struck by Saffir-Simpson category 4 or 5 hurricanes twice during the last 1200 yr, around 1170 and 860 cal yr BP, suggesting an annual landfall probability of 0.17% for these intense hurricanes. The charcoal data suggest that intense fires occurred after each of these hurricanes. The pollen data suggest that populations of halophytic plants (Chenopodiaceae) and heliophytic shrubs (Myrica) expanded after the hurricane strikes, probably due to saltwater intrusion into the marshes and soil salinization caused by overwash processes. Populations of pines (Pinus sp.) decreased significantly after each intense hurricane and the ensuing intense fire, suggesting that repeated hurricane-fire interactions resulted in high tree mortality and probably impeded recruitment and recovery. Our data support the hypothesis that the likelihood and intensity of fire increased significantly after a major hurricane, producing responses by vegetation that are more complex and unpredictable than if the disturbance agents were acting singly and independently.     

School vulnerability to disaster: examination of school closure,demographic, and exposure factors in Hurricane Ike’s wind swath

Damage and destruction to schools from climate-related disasters can have significant and lasting impacts on curriculum and educational programs, educational attainment, and future income-earning potential of affected students. As such, assessing the potential impact of hazards is crucial to the ability of individuals, households, and communities to respond to natural disasters, extreme events, and economic crises. Yet, few studies have focused on assessing the vulnerability of schools in coastal regions of the USA. Using Hurricane Ike’s tropical storm wind swath in the State of Texas as our study area, we: (1) assessed the spatial distribution patterns of school closures and (2) tested the relationship between school closure and vulnerability factors (namely physical exposure and school demographics) using zero-inflated negative binomial regression models. The regression results show that higher probabilities of hurricane strikes, more urbanized school districts, and school districts located in coastal counties on the right side of Ike’s path have significant positive associations with an increase in the number of school closure days. Socioeconomic characteristics were not significantly associated with the number of days closed, with the exception of proportion of Hispanic youth in schools, a result which is not supported by the social vulnerability literature. At a practical level, understanding how hurricanes may adversely impact schools is important for developing appropriate preparedness, mitigation, recovery, and adaptation strategies. For example, school districts on the right side of the hurricane track can plan in advance for potential damage and destruction. The ability of a community to respond to future natural disasters, extreme events, and economic crises depends in part on mitigating these adverse effects.     

How hurricane attributes determine the extent of environmental effects: Multiple hurricanes and different coastal systems

The most recent spate of hurricanes to strike the United States and Caribbean (1989 to the present) has occurred when many of the affected areas had long-term water quality and biological data collection efforts ongoing, as well as special follow-up studies. These data have allowed researchers to obtain a much clearer picture of how individual characteristics of hurricanes interact with human land use to lead to various types and degrees of environmental effects. Common deleterious water quality effects associated with hurricanes include excessive nutrient loading, algal blooms, elevated biochemical oxygen demand and subsequent hypoxia and anoxia, fish and invertebrate kills, aquatic animal displacements, large scale releases of chemical pollutants and debris from damaged human structures, exacerbated spread of exotic species and pathogens, and pollution of water with fecal microbial pathogens. These and other effects may or may not occur, or occur to varying degrees, depending upon individual hurricane characteristics including category, point of landfall, wind speed, amount of rainfall, and path after landfall. Landfall in a populous area, a post-landfall trajectory upriver toward a headwater region, passage along a floodplain containing pollution sources (such as wastewater treatment plants, concentrated animal feeding operations, and septic systems), and intensity sufficient to damage power generation will all lead to increased environmental damage. We suggest a number of recommendations for post-hurricane water sampling parameters and techniques, and provide several management-oriented recommendations for better coastal and floodplain land use aimed at lessening the water quality effects of hurricanes.     

Temporal variation of tropical cyclones in the North Atlantic Basin

A total of 269 tropical storms and hurricanes originated in the North Atlantic basin from 1960–1989. Of these, 76 made landfall on the continental United states. This study divides the 76 tropical storms into their month of formation. Seasonal shifts in the principal areas of tropical cyclone formation over the Atlantic basin have been recognized for many decades. The results of the study suggest that the early and late season tropical cyclones develop in areas which are first affected by the position of the sun, resulting in an increase in water temperatures. These cyclones normally make landfall along the Gulf Coast and usually are of low intensity. Formation areas shift eastward in mid-summer with a slight increase in intensity. By late August and early September, the formation areas have extended to the Cape Verde Islands. These storms tend to strike the east coast of the US and are normally more intense. By the end of the hurricane season, the primary formation area has shifted back to the Gulf of Mexico, with low intensity storms affecting the Gulf Coast.     

Hurricane storm surge simulations for Tampa Bay

Using a high resolution, three-dimensional, primitive equation, finite volume coastal ocean model with flooding and drying capabilities, supported by a merged bathymetric-topographic data set and driven by prototypical hurricane winds and atmospheric pressure fields, we investigated the storm surge responses for the Tampa Bay, Florida, vicinity and their sensitivities to point of landfall, direction and speed of approach, and intensity. All of these factors were found to be important. Flooding potential by wind stress and atmospheric pressure induced surge is significant for a category 2 hurricane and catastrophic for a category 4 hurricane. Tide, river, and wave effects are additive, making the potential for flood-induced damage even greater. Since storm surge sets up as a slope to the sea surface, the highest surge tends to occur over the upper reaches of the bay, Old Tampa Bay and Hillsborough Bay in particular. For point of landfall sensitivity, the worst case is when the hurricane center is positioned north of the bay mouth such that the maximum winds associated with the eye wall are at the bay mouth. Northerly (southerly) approaching storms yield larger (smaller) surges since the winds initially set up (set down) water level. As a hybrid between the landfall and direction sensitivity experiments, a storm transiting up the bay axis from southwest to northeast yields the smallest surge, debunking a misconception that this is the worst Tampa Bay flooding case. Hurricanes with slow (fast) translation speeds yield larger (smaller) surges within Tampa Bay due to the time required to redistribute mass.     

An analysis of coastal and inland fatalities in landfalling US hurricanes

Improvements in hurricane forecasts allowing for more timely evacuations from storm-surge zones are credited with reducing lethality of US landfalling hurricanes. The deadly reach of a hurricane, however, is not limited to storm-surge zones. About 80% of direct US hurricane fatalities since 1970 occurred outside of landfall counties, with most of these fatalities caused by inland flooding. We construct a geographic information system database combining the location and cause of fatalities, estimated wind speeds, and rainfall amounts for the entire track of the storm for landfalling US hurricanes between 1970 and 2007. We analyze the determinants of total fatalities and deaths due to freshwater drowning and wind. Inclusion of inland fatalities results in no downward trend in lethality over the period, in contrast to prior research. Local storm conditions significantly affect lethality, as one-inch and one-knot increases in rainfall and wind increase total fatalities by 28 and 4%. Rainfall significantly increases freshwater-drowning deaths and is insignificant for wind deaths, while the opposite relation holds for wind speed. While coastal counties do not exhibit a significantly higher amount of lethality risk versus inland counties for total or wind-driven fatalities, freshwater-drowning fatalities occur most frequently in inland counties along the center of the storm path and its outer county tiers as we have defined them.     

Satellite microwave detected SST anomalies and hurricane intensification

Sea surface temperature (SST) from the remotely sensed infrared measurements, like the GOES, AVHRR, and MODIS, etc., show missing values of SST over the cloudy regions associated with hurricanes. While satellite microwave measurements, like the Tropical Rainfall Measuring Mission (TRMM) microwave imager (TMI), can provide SST even under cloudy conditions. Both satellite microwave measurements and buoy observations show SST increase in advance of significant hurricane intensification. Moreover, hurricane intensification may also be related to the location of high SST. Our results indicate pre-existing high SST anomaly (SSTA) located at the right side of the storm track for Hurricane Katrina. Numerical simulations also confirm the important impacts of SSTA location on hurricane intensification. Similar situations are also found for Hurricanes Rita and Wilma. In contrast, if there is no high SSTA at the right location, hurricane may not undergo further intensification. This may explain why not all tropical cyclones associated with warm waters can attain peak intensity (categories 4 and 5) during their life cycle, and partially explains why hurricanes do not reach the maximum potential intensity as calculated only according to the magnitude of SST.     

Storms of tropical origin: a climatology for New York State,USA (1851–2005)

The tropical storm database used in this study was obtained from the National Oceanic and Atmospheric Administration’s (NOAA) Coastal Service Center, using the Historical Hurricane Tracks tool. Queries were used to determine the number of storms of tropical origin that have impacted the State and each of its counties. A total of 76 storms of tropical origin passed over New York State between 1851 and 2005. Of these storms, 14 were classified as hurricanes. The remaining hurricanes passed over New York State as weaker or modified systems—27 tropical storms, 7 tropical depressions, and 28 extratropical storms (ET). Long Island experiences a disproportionate number of hurricanes and tropical storms. The average frequency of hurricanes and storms of tropical origin (all types) is one in every 11 years and one in every 2 years, respectively. September is the month of greatest frequency for storms of tropical origin, although the storms of greatest intensity tend to arrive later in the hurricane season and follow different poleward tracks. While El Nino Southern Oscillation (ENSO) cycles appear to show some influence, the frequency and intensity of storms of tropical origin appear to follow a multidecadal cycle. Storm activity was greatest in both the late 19th and 20th centuries. During periods of increased storm frequency and intensity storms reached New York State at progressively later dates. While the number and timing of storms of tropical origin is likely to increase, this increase appears to be attributed to a multidecadal cycle, as opposed to a trend in global warming.     

Dissolved oxygen dynamics in Charlotte Harbor and its contributing watershed, in response to hurricanes Charley, Frances, and Jeanne—Impacts and recovery

On August 13, 2004, Hurricane Charley came ashore in the Charlotte Harbor watershed. Surface winds at the time of landfall were estimated at 130 knots. The track of the hurricane roughly followed the floodplain of the Peace River, causing massive defoliation and mortality of native vegetation and planted citrus groves, as well as substantial damage to human habitation and various infrastructure elements. Eight days after landfall, a water quality monitoring effort documented hypoxic (<2 mg I⁻¹) to nearly anaerobic (<0.5 mg I⁻¹) dissolved oxygen (DO) values throughout the vast majority of the Peace River's c. 6,000 km² watershed. Low DO values appeared to be related to high values of both dissolved organic matter and suspended materials. Hypoxic conditions in Charlotte Harbor itself, occurred within 2 wk of landfall. Approximately 3 wk after the landfall of Hurricane Charley, Hurricane Frances struck the east coast of Florida, causing further wind damage and bringing substantial amounts of rain to the Charlotte Harbor watershed. Three weeks later still, Hurricane Jeanne caused similar damage to the same area. In response to the combined effects of these three hurricanes, DO values in the Peace River did not recover to pre-hurricane levels until approximately 2–3 mo later. The spatial and temporal pattern of DO fluctuations appeared to be related to the proximity of sampling locations to the path of the eyewall of the first of the three hurricanes. Within the Harbor itself, the duration of hypoxic conditions was less than that recorded within the Peace River, perhaps reflecting greater dilution of oxygen-poor waters from the watershed with less-affected water from the Gulf of Mexico.     

A numerical study on rapid intensification of Hurricane Charley (2004) near landfall

The rapid intensification of Hurricane Charley (2004) near landfall is studied using the fifth-generation Pennsylvania State University/National Center for Atmospheric Research (PSU/NCAR) Mesoscale Model (MM5) and its adjoint system for both vortex initialization and forecasts. A significant improvement in both track and intensity forecasts is achieved after an ill-defined storm vortex, derived from large-scale analysis, in the initial condition is replaced by the vortex generated by a four-dimensional data variational (4D-Var) hurricane initialization scheme. Results from numerical experiments suggest that both the inclusion of the upper-level trough and the use of high horizontal resolution (6 km) are important for numerical simulations to capture the observed rapid intensification as well as the size reduction during the rapid intensification of Hurricane Charley. The approach of the upper-level trough significantly enhanced the upper-level divergence and vertical motion within simulated hurricanes. Small-scale features that are not resolvable at 18 km resolution are important to the rapid intensification and shrinking of Hurricane Charley (2004). Numerical results from this study further confirm that the theoretical relationship between the intensification and shrinking of tropical cyclones based on the angular momentum conservation and the cyclostrophic approximation can be applied to the azimuthal mean flows.     

Experimental superensemble forecasts of tropical cyclones over the Bay of Bengal

This study entails the implementation of an experimental real time forecast capability for tropical cyclones over the Bay of Bengal basin of North Indian Ocean. This work is being built on the experience gained from a number of recent studies using the concept of superensemble developed at the Florida State University (FSU). Real time hurricane forecasts are one of the major components of superensemble modeling at FSU. The superensemble approach of training followed by real time forecasts produces the best forecasts for tracks and intensity (up to 5 days) of Atlantic hurricanes and Pacific typhoons. Improvements in track forecasts of about 25–35% compared to current operational forecast models has been noted over the Atlantic Ocean basin. The intensity forecasts for hurricanes are only marginally better than the best models. In this paper, we address tropical cyclone forecasts over the Bay of Bengal for the years 1996–2000. The main result from this study is that the position and intensity errors for tropical cyclone forecasts over the Bay of Bengal from the multimodel superensemble are generally less than those of all of the participating models during 1- to 3-day forecasts. Some of the major tropical cyclones, such as the November 1996 Andhra Pradesh cyclone and October 1999 Orissa super cyclone were well handled by this superensemble approach. A conclusion from this study is that the proposed approach may be a viable way to construct improved forecasts of Bay of Bengal tropical cyclone positions and intensity.     

Resilience: a capacity and a myth: findings from an in-depth case study in disaster management research

Hurricane Andrew, which made landfall on August 24, 1992, was one of the most destructive hurricanes in American history, causing atypically high levels of psychological and physical health impairment among the resident population and especially among vulnerable groups. This article investigates whether maternal exposure to Hurricane Andrew during pregnancy increased the risk of dystocia (or dysfunctional labor) and infant delivery by cesarean section, the standard medical response to abnormal labor progression. We analyze 297,996 birth events in Miami-Dade and Broward counties in Florida from 1992 to 1993 using propensity score methodology with stratification and nearest-neighbor matching algorithms. Results show that hurricane-exposed pregnant women were significantly more likely to experience stress-induced abnormal labor and cesarean delivery outcomes as compared to statistically matched comparison groups. The conclusion details the policy implications of our results, with particular attention to the importance of maternal prenatal care in the aftermath of disasters.     

Characteristics of severe Atlantic hurricanes in the United States: 1949–2006

Property insurance data available for 1949–2006 were assessed to get definitive measures of hurricane losses in the U.S. Catastrophes, events causing >$1 million in losses, were most frequent in the Southeast and South climate regions. Losses in these two regions totaled $127 billion, 85% of the nation’s total losses. During the period 1949–2006 there were 79 hurricane catastrophes, causing $150.6 billion in losses and averaging $2.6 billion per year. All aspects of these hurricanes showed increases in post-1990 years. Sizes of loss areas averaged one state in 1949–1967, but grew to 3 states during 1990–2006. Seven of the ten most damaging hurricanes came in 2004 (4) and 2005 (3). The number of hurricanes also peaked during 1984–2006, increasing from an annual average of 1.2 during 1949–1983 to 2.1 per year. Losses were $49.3 billion in 1991–2006, 32% of the 58-year total. Various reasons have been offered for such recent increases in hurricane losses including more hurricanes, more intense tropical storms, increased societal vulnerability in storm-prone areas, and a change in climate due to global warming, although this is debatable.     

Agent-Based Modeling and Analysis of Hurricane Evacuation Procedures for the Florida Keys

The unique geography of the Florida Keys presents both high risk of hurricane landfall and exceptional vulnerability to the effects of a hurricane strike. Inadequate hurricane shelters in the Keys make evacuation the only option for most residents, but the sole access road can become impassable well in advance of a major storm. These extraordinary conditions create challenges for emergency managers who must ensure that appropriate emergency plans are in place and to ensure that an orderly exodus can occur without stranding large numbers of people along an evacuation route with inadequate shelter capacity. This study attempts to answer two questions: (1) What is the minimum clearance time needed to evacuate all residents participating in an evacuation of the Florida Keys in advance of a major hurricane for 92,596 people – a population size calculated based on the 2000 US Census population data, census undercounts, and the number of tourists estimated to be in the area? (2) If a hurricane makes landfall in the Keys while the evacuation is in progress, how many residents will need to be accommodated if the evacuation route becomes impassable? The authors conducted agent-based microsimulations to answer the questions. Simulation results suggest that it takes 20 h and 11 min to 20 h and 14 min to evacuate the 92,596 people. This clearance time is less than the Florida state mandated 24-h clearance time limit. If one assumes that people evacuate in a 48-h period and the traffic flow from the Keys would follow that observed in the evacuation from Hurricane Georges, then a total of 460 people may be stranded if the evacuation route becomes impassable 48 h after an evacuation order is issued. If the evacuation route becomes impassable 40 h after an evacuation order is issued, then 14,000 people may be stranded.     

Hurricanes 2004: An overview of their characteristics and coastal change

Four hurricanes battered the state of Florida during 2004, the most affecting any state since Texas endured four in 1884. Each of the storms changed the coast differently. Average shoreline change within the right front quadrant of hurricane force winds varied from 1 m of shoreline advance to 20 m of retreat, whereas average sand volume change varied from 11 to 66 m³ m⁻¹ of net loss (erosion). These changes did not scale simply with hurricane intensity as described by the Saffir-Simpson Hurricane Scale. The strongest storm of the season, category 4 Hurricane Charley, had the least shoreline retreat. This was likely because of other factors like the storm's rapid forward speed and small size that generated a lower storm surge than expected. Two of the storms, Hurricanes Frances and Jeanne, affected nearly the same area on the Florida east coast just 3 wk apart. The first storm, Frances, although weaker than the second, caused greater shoreline retreat and sand volume erosion. As a consequence, Hurricane Frances may have stripped away protective beach and exposed dunes to direct wave attack during Jeanne, although there was significant dune erosion during both storms. The maximum shoreline change for all four hurricanes occurred during Ivan on the coasts of eastern Alabama and the Florida Panhandle. The net volume change across a barrier island within the Ivan impact zone approached zero because of massive overwash that approximately balanced erosion of the beach. These data from the 2004 hurricane season will prove useful in developing new ways to scale and predict coastal-change effects during hurricanes.     

Yucatan after the wind: Human and environmental impact of hurricane Gilbert in the central and Eastern Yucatan Peninsula

Hurricane Gilbert made its landfall on the Yucatan peninsula on September 14, 1988, destroying valuable resort property in Cancun, Isla Mujeres, and Cozumel. Salt flats (salinas) in northern Quintana Roo and Yucatan were flooded by the storm surge and coastal marine ecosystem were devastated. Inland, the hurricane caused damages to houses, power lines, and farming land, but increased opportunities for garden hunting on destroyed fields. Twenty percent of tropical rain forest suffered losses of canopy and in the deciduous forests of north-central Yucatan noxiuous insects populations increased. Although damaging in its ultimate effects, hurricane Gilbert tested the resilience of Maya Indians and their readiness for post-disaster accommodation.     

Impact and response of southwest Florida mangroves to the 2004 hurricane season

Although hurricane disturbance is a natural occurrence in mangrove forests, the effect of widespread human alterations on the resiliency of estuarine habitats is unknown. The resiliency of mangrove forests in southwest Florida to the 2004 hurricane season was evaluated by determining the immediate response of mangroves to a catastrophic hurricane in areas with restricted and unrestricted tidal connections. The landfall of Hurricane Charley, a category 4 storm, left pronounced disturbances to mangrove forests on southwest Florida barrier islands. A significant and negative relationship between canopy loss and distance from the eyewall was observed. While a species-specific response to the hurricane was expected, no significant differences were found among species in the size of severely impacted trees. In the region farthest from the eyewall, increases in canopy density indicated that refoliation and recovery occurred relatively quickly. There were no increases or decreases in canopy density in regions closer to the eyewall where there were complete losses of crown structures. In pre-hurricane surveys, plots located in areas of management concern (i.e., restricted connection) had significantly lower stem diameter at breast height and higher stem densities than plots with unrestricted connection. These differences partially dictated the severity of effect from the hurricane. There were also significantly lower red mangrove (Rhizophora mangle) seedling densities in plots with restricted connections. These observations suggest that delays in forest recovery are possible in severely impacted areas if either the delivery of propagules or the production of seedlings is reduced by habitat fragmentation.     

A surge response function approach to coastal hazard assessment. Part 2: Quantification of spatial attributes of response functions

In response to the 2004 and 2005 hurricane seasons, surge risk assessment approaches have been re-evaluated to develop more rapid, reliable methods for predicting the risk associated with extreme hurricanes. Here, the development of dimensionless surge response functions relating surge to hurricane meteorological parameters is presented. Such response functions present an opportunity to maximize surge data usage and to improve statistical estimates of surge probability by providing a means for defining continuous probability density functions. A numerical modeling investigation was carried out for the Texas, USA coastline to develop physical scaling laws relating storm surge response with hurricane parameters including storm size, intensity, and track. It will be shown that these scaling laws successfully estimate the surge response at any arbitrary location for any arbitrary storm track within the study region. Such a prediction methodology has the potential to decrease numerical computation requirements by 75% for hurricane risk assessment studies.