HIV screening is recommended to destigmatize the condition, prevent partner transmission, and postpone AIDS progression. However, determinants associated with implementation of opt-out HIV screening are not well understood. The purpose of the study was to examine determinants that predicted odds of HIV screening for persons impacted by Hurricane Sandy, and how these factors differed according to demographic characteristics, geographical attributes, health-related quality of life score, access-to-care, and health insurance status. The social ecologic model provided the framework for this multilevel cross-sectional study that included New Jersey data from the Behavior Risk Factor Surveillance System. Bivariate chi-square, simple logistic regression, and adjusted multivariate and weighted logistic regression analyses were performed to estimate HIV screening odds. Findings indicated a significant odds ratio with access to care post- Hurricane Sandy and HIV screening (odds ratio = 1.74, 95% CI = 1.38-2.21). The positive social change implications may include assisting people to develop realistic plans for HIV screening, improving understanding of HIV screening determinants, and raising awareness of the risk factors related with access to medical care post-Hurricane Sandy.

This dissertation consists of three papers in environmental and natural resource economics. The first paper estimates the value of statistical lives (VSL) from hurricane evacuation behavior through an empirical analysis. I present empirical models that predict individuals' willingness to pay (WTP) for avoiding hurricane risks revealed through their evacuation behavior. Using survey data from Texas residents (who were affected by Hurricane Ike), I analyze the individuals’ hurricane evacuation decisions and their corresponding WTP for evacuation. I also estimate the individuals' WTP for avoiding hurricane risks under both voluntary and mandatory evacuation orders and calculate the associated VSL. The findings can be useful to emergency management agencies for evacuation planning. In the second paper, I study market responses to multiple hurricanes based on evidence from real estate sales data. Unlike earlier studies that examined the effect of hurricane exposures on property value, the present study considers how multiple hurricane hits affect the home value. I use repeat sales data from three counties in Florida from 2000 to 2010 and develop a hedonic price model. The findings identify the determinants that influence the property value and provide valuable insights for homebuyers and sellers. The study also provides useful insights regarding the benefits of hurricane mitigations to Florida residents and beyond. The third paper investigates the time preference and the dynamics of evacuation behavior based on evidence from Hurricane Ike and Hurricane Sandy. This paper contributes to the literature on households’ evacuation timing decisions by investigating the factors influencing people’s time preference for evacuation behavior. Unlike other studies, I examine the residents’ evacuation behavior across the Gulf coast as well as the Northeast and Mid-Atlantic coasts from a comparative perspective. I use one survey dataset from Texas residents who experienced Hurricane Ike and another survey dataset from the Northeastern and Mid-Atlantic US states that were affected by Hurricane Sandy. The results provide insights for future hurricane evacuation planning and emergency management.

acase@tulane.edu / 1 / Melissa Schigoda

Hurricane Katrina made landfall on the American Gulf Coast on August 29, 2005. Biloxi, Mississippi, a small town on the coast, was one of the towns devastated by the storm. A decade after the storm, recovery remains an ongoing process. My ethnographic research in 2006, 2010, and 2011 and media and historic document analysis throughout these ten years explore this recovery process and what pre-disaster cultural, social, political, and economic issues have shaped Biloxi and Biloxians' recovery. The small coastal city of Biloxi sits on the Mississippi Sound of the Gulf of Mexico. The city's history and residents' identities are intertwined with this waterfront location. Biloxians rely on the Gulf for recreation and job opportunities, particularly in the long-standing seafood and tourism industries. Scattered piers are filled with recreational and shrimping boats. Casinos dot the shoreline where seafood processing plants once stood. Many Biloxians still proudly identify with the city's coastal location, neighborhoods they were raised in or lived in before Katrina, their perceived socioeconomic class status, and their own and their ancestor's racial, ethnic, and national identities. However, Biloxi's waterfront location also makes the city prone to hurricane strikes. Historic storms like Camille in 1969 and Fort Lauderdale in 1947 have affected the city's development and influenced residents' beliefs and behaviors during their preparation for Katrina. Biloxians were aware of Katrina's predicted landfall in the days and hours before the storm, but this history of hurricanes influenced many residents' decisions to remain in the city for the storm. Many residents I spoke to described their belief that survival in previous storms indicated they would survive Katrina. Other pre-Katrina processes influenced Biloxians' preparations for, coping with, and response to the disaster, as expected in vulnerability theory. Poorer and working class residents were less able to prepare for or evacuate before the storm, if they chose to do so. Residents in higher risk neighborhoods like East Biloxi found themselves affected more severely by the storm, often losing much of their homes and lives. Biloxians' with less political and economic power struggled to keep their voice heard as city and other government officials laid the framework for recovery. Pre-Katrina Biloxians' cultural, political, and economic inequalities directly affected the recovery process. To better understand these influences, in this research I use a political economy approach to describe and analyze Biloxi's recovery from Katrina. To strengthen this analysis, I have also drawn on theories regarding vulnerability and resilience, risk and uncertainty, and cultural-historical context. Each of these approaches contributes to a better understanding of how post-disaster recovery processes work - particularly in the case of post-Katrina Biloxi. This work also builds on disaster anthropology and social science research that rejects the concept of disasters as isolated events and instead argues that disasters are influenced by broader and long-standing cultural, political, and economic processes. In this work I also bring this argument for a holistic approach into long-term disaster recovery. The holistic anthropological approach to the post-Katrina Biloxi that I have used here reveals the importance of understanding a range of facts and processes that exist before, during, and after a disaster to explore the recovery process. Post-Katrina Biloxi is as much a product of pre-Katrina Biloxi as it is a product of the effects of the hurricane itself.

The beach profiles pre-and post-the Hurricane Hermine (2016) and Irma (2017) along the Sand Key barrier island were collected to quantify longshore variations in storm induced beach changes as well as to compare the beach changes caused by hydrodynamic conditions of the two different hurricanes. Cross-shore beach profile are examined in 4 sections including dune field, dry beach, sand bar and whole beach to calculate beach change. The volume change for each section and shoreline contour change before and post the hurricane was computed. Hydrodynamic conditions were obtained from adjacent NOAA’s tide and wave gauges. Both hurricanes generated high offshore waves, with Hurricane Hermine generated waves mostly from southwest, and Irma generated waves dominantly from northeast. Hurricane Hermine generated a storm surge of up to 1 m. While hurricane Irma generated negative surge of -1.1 m. Several beach profile parameters such as the foreshore slope, as well as volume changes of dune field, dry beach and sand bar induced by the two hurricanes were computed. Under both storms, the foreshore slope became steeper after the storm north of the headland, while the foreshore slope became gentler south of the headland. Storm surge plays an important role in inducing beach erosion. Hurricane Hermine with 1 m surge caused significant dune erosion in terms of dune volume loss and dune line retreat. On the other hand, hurricane Irma with negative surge only caused minor dune erosion. Sand bar moved seaward during both hurricanes, with Irma induced a much greater offshore movement than that of Hermine. In addition, the sand bar height decreased significantly during Irma. In contrast, during Hermine the sand bar height remained largely similar before and after the storm. Large alongshore variations in beach erosion was observed during both hurricanes as influenced by background erosion rate and direction of incident waves as they approaching the curved shoreline. For both storms, the erosional hot spot at North Sand Key with the highest background erosion rate suffered the most sand loss over the entire profile. More sand was eroded from the dry beach along the broad headland than along the beaches both north and south of it. Corresponding to the higher volume of dry beach erosion, shoreline retreat was also the largest around the headland. During Hurricane Hermine, the headland sheltering of the southerly approaching waves resulted in more erosion to the south than to the north. The opposite happened during Hurricane Irma with northerly approaching wave. More erosion occurred to the north of the headland than that to the south. Systematic measurement of beach profile beach and after hurricanes can improve our understanding on beach morphodynamics on storm induced beach changes.

This study examines the impact of natural climate variability and long-term climate change on North Atlantic tropical cyclone (TC) tracks. Using data from HURDAT for the period 1950-2007, we categorize Atlantic TCs that form in the Main Development Region into one of three track types: Straight-Moving (SM), Recurving Landfall (RCL), and Recurving Ocean (RCO) TCs. As expected, the SM storms are associated with a westward extension and strengthening of the North Atlantic Subtropical High (NASH) whereas the RCO storms coincide with a weakening of the NASH. The presence of El Nino conditions in the tropical Pacific is found to be associated with a weakening and eastward retreat of the NASH, an increase in the percentage of RCO TCs, and a decrease in the percentage of RCL TCs. Using 6-hourly wind fields from NCEP-NCAR Reanalyses, simulated tracks are computed for each historical storm in the sample using the Beta Advection Model (BAM). Using observed genesis locations, the BAM successfully reproduces the differences in TC tracks between SM, RCO and RCL storm types. When storm genesis is uniformly distributed over the MDR we find that RCL and RCO storms still exhibit a distinct difference in tracks, suggesting that differences in the large-scale steering flow over the tropical Atlantic are primarily responsible for their track differences. However the SM TCs exhibit a more northward track under the uniform genesis experiment, indicating that the more southern and western genesis location of these storms is an important contributor in determining their tracks. The observed difference between TC tracks during El Nino and La Nina events is also reproduced by the BAM under both observed and uniformly seeded genesis experiments, suggesting that it is the changes in the large-scale steering flow over the Atlantic that is responsible the larger percentage of RCO storm tracks during El Nino events. The influence of anthropogenic warming is examined using a 7 member ensemble comparing the 2xCO2 experiment to a pre-industrial control. Increased sea level pressure over the northeast and northwest quadrants of the Atlantic does not alter the average TC track.

The tropical cyclone (TC) and environmental interaction is not fully understood. Previous studies have demonstrated that this interaction affects intensity change. The studies found that intensification is favored in low shear, moist environments, with high sea surface temperatures (SST). However, little precise quantification was provided, especially in terms of the impact of environmental water vapor on TC intensity change. This work addresses the TC interaction with the environmental water vapor. Results from a comprehensive statistical study show that TC intensification is more likely to occur in an anomalously moist environment than a dry environment. However, only a small amount of the total variance is explained. When assessing the effect of vertical wind shear along with environmental water vapor, more of the variance is explained. Water vapor not only affects TC intensity. Prior modeling studies have demonstrated impacts from environmental water vapor on TC structure. These impacts can also affect intensity change. Specifically, enhanced water vapor content within the TC enhances the rainbands, which can lead to an eyewall replacement cycle, causing a temporary weakening, followed by re-intensification. This thesis evaluates observational and high resolution MM5 model output from Hurricanes Katrina and Rita from the Hurricane Rainband and Intensity Experiment (RAINEX) to evaluate the effects of varying water vapor distributions on TC structure. While the two hurricanes were of similar intensity, they had different water vapor distributions and structures. Rita underwent an eyewall replacement cycle while under RAINEX surveillance while Katrina did not. Rita was also located within a dry environment and had a strong horizontal moisture gradient, while Katrina was in a moist environment and had a weak moisture gradient. Results suggest that a strong horizontal water vapor gradient, with a moist TC and dry outer environment may confine the hurricanes into a pattern that causes them to have high circularity, promoting the formation of a secondary eyewall. The dry outer environment had strong atmospheric stability and was less favorable for deep convection far from the center in the Rita case. The moist environment in the Katrina case was more unstable. This may have allowed for the rainbands to be farther from the center in a less circular pattern than Rita. The results presented in this thesis suggest that this pattern is less favorable for an eyewall replacement cycle.

A new oceanic climatology to calculate ocean heat content (OHC) was developed for application year-round in the Atlantic Ocean basin. The Systematically Merged Atlantic Regional Temperature and Salinity (SMARTS) Climatology blends temperature and salinity fields from the World Ocean Atlas 2001 (WOA) and Generalized Digital Environmental Model v.3.0 (GDEM) at 1/4° resolution. This higher resolution climatology better resolves features in the Gulf of Mexico (GOM), including the Loop Current and eddy structures, than the previous coarser 1/2° products. Daily mean isotherm depths of the 20° C (D20) and 26° C (D26) (and their mean ratio), reduced gravity (e.g., 2-layer model), mixed layer depth (MLD), and OHC were estimated from the blended climatology. Using SMARTS with satellite-derived surface height anomaly and SST fields, daily values of D20, D26, MLD, and OHC were calculated from 1998 to 2010 using a two-layer model approach. Airborne and ship-deployed eXpendable BathyThermographs (XBT), long-term moorings, and Argo profiling floats provided the in-situ data to blend and assess the SMARTS Climatology. A clear, direct relationship emerged from the detailed analysis between satellite-derived and in-situ measurements of isotherm depths and OHC. This new climatological approach created a more accurate estimation of isotherm depths and OHC from satellite radar altimetry measurements, which can be used in hurricane intensity forecasts from the Statistical Hurricane Intensity Prediction Scheme (SHIPS). The Mainelli (2000) technique of calculating OHC was reexamined to most accurately project sea surface height anomalies (SSHA) into changes in depths of D20, D26, and MLD. SSHA surface features were tracked to determine realistic drift velocities ingested into the objective analysis algorithm. The former OHC algorithm assumed a climatological MLD, however observations show large temporal variability of MLD. Using a SSHA-dependent MLD for the OHC estimation improves the two-layer model by 5%. Upper ocean thermal structure estimations improved by 25% using the SMARTS Climatology as compared to that of Mainelli (2000).

Accurate prediction of Tropical Cyclones (TC) is vital for the protection of life and property in areas that are prone to their destructive forces. While significant improvements have been made in forecasting TC track, intensity remains a challenge. It is hypothesized that accurate TC intensity forecast requires, among other things, an adequate initial description of their inner-core region. Therefore, there must be reliable observations of the inner-core area of the TC and effective data assimilation (DA) methods to ingest these data into the Numerical Weather Prediction (NWP) models. However, these requirements are seldom met at the relatively low resolution of operational global prediction models and the lack of routine observations assimilated in the TC inner-core. This study tests the impacts of assimilating inner-core Airborne Doppler Radar (ADR) winds on the initial structure and subsequent intensity forecast of Hurricane Ike (2008). The 4-dimensional variational (4DVar) and the 3-dimensional variational (3DVar) methods are used to perform DA while the Weather Research and Forecasting (WRF) model is used to perform forecasts. It is found that assimilating data helps to initialize a more realistic inner-core structure using both DA methods. Additionally, the resulting short-term and long-term intensity forecasts are more accurate when data is assimilated versus cases when there is no DA. Additionally, it is found that in some cases the impact of DA lasts up to 12 hours longer with 4DVar versus 3DVar. It is shown that this is because the flow-dependent 4DVar method produces more dynamically and balanced analysis increments compared to the static and isotropic increments of 3DVar. However, the impact of using both methods is minimal in the long-range. The analyses show that at longer forecast range the dynamics of hurricane Ike was influenced more by outer environment features than the inner-core winds.

Thesis (M.A.)--University of North Carolina at Wilmington, 2006. / Includes bibliographical references (leaves: 66-66)