Peer Reviewed Papers on Hurricane Katrina Environmental Impacts
Introduction
Over the past several decades, the increase in intensity and frequency of natural hazards, such equally hurricanes (Goldenberg et al., 2001; Webster et al., 2005), prolonged drought and heat waves (Meehl and Tebaldi, 2004) has brought significant impacts on social-ecological systems. In 2005, Hurricane Katrina impacted an estimated two,331 square kilometers, flooded over 80% of the city of New Orleans, and displaced 400,000 people causing significant outmigration (Lewis et al., 2017). The scale and degree of devastation and population relocation from this natural risk exceeded the impacts from previous events such as the 1900 Galveston Hurricane, the 1906 San Francisco earthquake and the 1927 Mississippi Flood (Elliott and Pais, 2006). Information technology is well-documented that Hurricane Katrina'southward devastation was widespread, all the same, at that place has been less give-and-take and exploration on the equity of the impacts and recovery using a social-ecological resilience lens.
Chance enquiry has grown significantly in the past decades across diverse fields of study, with enquiry questions primarily associated with vulnerability, recovery and resilience (Opdyke et al., 2017). Vulnerability is defined as the characteristics of an private or a group that influence their capacity to anticipate, respond to, cope with, resist, and recover from an external disturbance (eastward.k., natural take chances) and the subsequent bear on on their livelihood and well-existence (Kelly and Adger, 2000; Wisner, 2004). It tin too be defined equally "the degree to which a system or system component is likely to experience harm due to exposure to a hazard, perturbation or stress" (Turner II et al., 2003). The assessment of vulnerability to natural hazards typically uses surrogate variables, representing sensitivity and adaptive capacities of an exposed entity, to construct a quantitative and comparable alphabetize amidst spatial units. The assessment tin serve as a baseline or context for conclusion and action that facilitates and informs prevention, planning, mitigation, adaptation, and recovery from the touch of natural hazards (Kelly and Adger, 2000; Cutter and Finch, 2008). An instance is the social vulnerability alphabetize (SoVI) to natural hazards developed by Cutter et al. (2003). The index aggregates demographic, housing, and neighborhood variables from the U.South. Census Bureau to examine the social vulnerability of U.S. counties to environmental hazards.
In contempo years, the discourse on natural hazards in both inquiry and policy domains, is shifting from vulnerability to resilience among U.S. federal agencies (Cutter et al., 2008, 2014; United states Department of Housing Urban Development, 2014) and on the international stage (Fekete, 2009; World Bank, 2013). Part of the reason for this shift is that resilience is more proactive and dynamic (Cutter et al., 2008). A challenge of assessing disaster resilience emerges when there is neither a single definition for disaster resilience nor a widely accepted way to measure it. Though in that location take been attempts to assess some dimensions of community resilience to natural disasters, there is still a lack of consistent and standard metrics or surrogate variables to evaluate disaster resilience in communities. The concept of resilience (from ecology) is described as "a measure of the persistence of systems and of their power to absorb alter and disturbance and still maintain the same relationship between populations or state variables" (Holling, 1973). Ecological resilience focuses on the ability of a system to withstand change and retain its processes and structures without shifting to a new regime, or an alternative state. Thus, resilience tin be measured by the magnitude of disturbance the system tin can tolerate and still persist (Carpenter et al., 2001). Ecological resilience was introduced to the social sciences in the context of vulnerability to climate furnishings (Timmerman, 1981; Cutter et al., 2003). Cutter et al. (2008) later on presented a disaster resilience framework at the community level. Heavily influenced past the piece of work of social vulnerability to environmental hazards, the enquiry attempts to appraise disaster resilience by applying an anterior method to explicate susceptibility through select variables intended to characterize the exposed entity. Cutter et al. (2008) argued that the conceptual frameworks of disaster resilience and vulnerability to natural hazards are both dynamic; nevertheless, the assessment presented was however static (i.e., evaluated but i snapshot at a time), which is very different from the quantitative approaches for assessing ecological resilience (Angeler et al., 2016). In the context of natural hazards, resilience is sometimes divers from an applied science perspective. An case is the National Ocean Service at the National Oceanic and Atmospheric Administration (NOAA, 2018) defining coastal resilience as "building the ability of a customs to bounce back after hazardous events such as hurricanes, coastal storms, and flooding, rather than simply reacting to impacts." Inside this context, many natural hazard studies focus on engineered and homo systems, including loss prevention and mail-disaster actions and planning to minimize disaster impacts (Bruneau et al., 2003; Cutter et al., 2008). An overemphasis upon engineering resilience (which assumes a single-state landscape) limits the agreement of a system's emergent capacities to withstand disruptions that are unforeseeable (Sikula et al., 2015) or the possibility that recovery to a previous country is impossible; hence, new regimes, processes and structures emerge (Chuang et al., 2018).
Vulnerability and resilience share some commonalities. For instance, the level of vulnerability tin bear upon the degree of resilience in a coupled human and natural arrangement (Turner II et al., 2003). There have been attempts to distinguish the two terms from diverse (and some contested) definitions (Turner 2 et al., 2003; Gallopín, 2006; Cutter et al., 2008). With the lens of sociology, Gotham and Campanella (2011) argue that resilience studies attempt to find "how and under what conditions ecological and homo communities conform and suit, or transform and introduce in response to a daze or traumatic event," and vulnerability studies investigate the root of hazards within coupled systems (e.g., Eakin et al., 2009), examine their varying capacities to respond to these hazards, and explore the "co-beingness of adaptive and maladaptive couplings in vulnerable systems" (Gotham and Campanella, 2011). In the context of hazards and disasters, Cutter et al. (2008) differentiated the two concepts by the temporal flow of the assessment. They described vulnerability as the "pre-event, inherent characteristics or qualities of social systems that create the potential for damage (Cutter et al., 2008)," while defining resilience as "the ability of a social system to respond and recover from disasters. Accordingly, the Cutter et al. (2008) definition of resilience includes those inherent conditions that allow the system to absorb impacts and cope with an outcome, as well every bit, mail service-event adaptive processes that facilitate the power of the social organization to re-organize, change, learn and answer to a threat."
Disaster resilience studies oftentimes address a fix of capacities and strategies for disaster readiness (Norris et al., 2008) which relates the ability of a customs to prepare and program for, absorb, recover from, and arrange to adverse events in a timely and efficient manner. Opdyke et al. (2017) documented the hazard studies from 1990 to 2015. Amongst the methods used for quantifying resilience to natural hazards, the top three approaches are (one) modeling or simulation (31% out of 241 studies), (2) literature review or theoretical framework (24%), and (3) GIS analysis (fifteen%). Specifically, the majority of the first blazon of inquiry focuses on economic aspects of resilience (i.east., economic modeling). The 2d type of research is largely narrative-based, conceptually developing a framework and indices to measure out resilience to natural hazards or climate events. For instance, Summers et al. (2017) presented a conceptual model that characterizes resilience to climate events from the natural environment, lodge, the built environment and risk domains. Such approaches utilize indicators (metrics) to establish a quantifiable baseline condition for comparing the degree of resilience to monitor progress. However, many of these disaster indicator studies are designed with generic goals, similar community resilience, implying or attempting to narrate an entire community's resilience to all hazards without specifying a detail threat, disturbance or unique characteristics of the exposed organization. Such a generalized cess could limit the awarding and utility of the approaches for exploring particular cases. This is in contrast to the Carpenter et al. (2001) argument that the nigh important step is to identify resilience "of what to what" considering unlike disturbances could touch a system in many ways and with dissimilar magnitudes and, building the chapters for resilience in one area may create vulnerabilities in other areas. Further, when evaluating resilience, ane must be enlightened of the existence of multiple regimes and understand that returning to a particular authorities (set of conditions) may be incommunicable (Carpenter et al., 2001).
There is extensive research on natural hazards but trivial has focused on the process of reconstruction (Kates et al., 2006), change and reorganization of social-ecological systems before and after a natural hazard. Although disturbances, such every bit natural hazards, often wreak havoc on human and natural systems, in the aftermath (reorganization phase), there is the potential opportunity for innovation, development and transformation (Folke, 2006, 2016). Panarchy describes the changing stages of complex adaptive systems equally they continually organize and construction within and across scales of space and time, and is characterized by a set of interconnected adaptive cycles (Gunderson and Holling, 2002). The four phases of adaptive cycles are release (or plummet), reorganization, exploitation, and conservation (Gunderson and Holling, 2002). The release/collapse stage is triggered by a disturbance that is large enough to change the land of a system. Reorganization describes the menses later on a disturbance where the system goes into an unstable period and has relatively low resistance to new innovations. Exploitation refers to a rapid growth (van der Leeuw, 2013) and exploitation of resource by organisation components (Gunderson and Holling, 2002). Lastly, conservation is the phase where organisation components gradually become more established and connected (van der Leeuw, 2013). In this land, the organisation becomes rigid and increasingly sensitive to disturbances (Allen et al., 2014). Disturbances have the potential to create opportunities for innovation, development (Folke, 2006), comeback, and beneficial transformation.
The Metropolis of New Orleans has rebuilt after previous natural disasters with the hope of emerging in a safer and more equitable way (Kates et al., 2006). This research aims to examine if this goal was reached by studying how social and ecological condition changed over time, while viewing Hurricane Katrina as a perturbation of the system. The effort focuses on the process of modify and reorganization after a natural hazard and applies the concept of ecological resilience in the analyses. Information technology takes an integrated approach that includes assessments from socio-demographic, economical, and environmental dimensions, to empathise the changes in this coastal city, before and after Hurricane Katrina. Specifically, we asked the post-obit questions: What are the impacts on human and natural systems in New Orleans, several years after Hurricane Katrina? Farther, Hurricane Katrina offered a window of opportunity to transform the organization. How has the arrangement changed over time? Extending the traditional approaches, nosotros used indicators to characterize the status of the organization. Also, rather than a snapshot, we introduce a dynamic component by assessing the status of New Orleans and southern Louisiana pre- and post-disaster using multiple fourth dimension steps. Nosotros besides incorporated ecological components by assessing land cover changes over time and incorporated breeding bird survey data every bit a proxy for evaluating ecosystem variation. Further, we evaluated these social-ecological changes using principal component analysis (PCA) and rather than only quantifying and comparing indicator variables, we used GIS to capture spatial patterns and observed the quality of alter by examining specific traits of the system. Finally, we discussed the limitations of our work and offer guidance on improving disaster resilience enquiry in the future.
Study Surface area
New Orleans is an archetype for coastal ecosystems under immediate threat by natural hazards and environmental change, specifically due to sea level rise and hurricanes (Gotham et al., 2014). The City is about 169 kilometers north of the Gulf of Mexico, located between the Mississippi River and Lake Pontchartrain. A large part of the metropolis is beneath sea level (lowest topography reaches 3–5 meters below bounding main level) (Dixon et al., 2006) and flooding of levees and floodwalls is expected when a storm reaches Category three or higher up (Carter, 2005). Prior to Katrina, the urban center'southward population declined from a meridian at 627,000 in 1960 to about 484,000 people in 2000 (United states of america Demography Bureau, 2017). In 2005, Hurricane Katrina resulted in more than 1500 deaths and 76.eight% of the population suffered from flooding. Correct afterward Hurricane Katrina, the population size of the city sharply decreased to 208,000 people in 2006, resulting in more than than a 57% reduction in population in the city when compared with the population of the city in 2000. In the post-Katrina era, the metropolis has revived and is experiencing increased wages and higher median household income, population growth (391,000 in 2016) and growing entrepreneurship (Liu and Plyer, 2010). Along with economic and population growth, there are also spikes in housing costs and offense rates, resulting in neighborhood instability and social conflicts (Gotham and Campanella, 2011). Our aim is to report the changes in this social-ecological arrangement over time.
Methods and Information Analysis
To better understand a organisation's capacity to withstand and adapt to natural hazards, the evaluation should examine the degree of pre-disturbance vulnerability or risk to the arrangement, and level of postal service-disturbance renewal, reorganization, and innovation (Gotham and Campanella, 2011). Thus, nosotros assessed and compared the weather before and after Hurricane Katrina through an integrated arroyo to interpret our findings in a disaster resilience context. Our assessment comprises the three core dimensions of coupled human being and natural systems: (ane) economic; (2) social; and (3) environmental. We examined the characteristics of the system temporally and spatially using representative variables from economic and social-demographic dimensions. In the environmental dimension, we examined the restoration and amending of the natural arrangement by evaluating land-encompass change and bird diversity over fourth dimension.
Since continuous data were not bachelor for all the social and economical variables, the assessment of before and after conditions comprises iii-time steps, year 2000 (pre-Katrina), 2009 (post-Katrina), and 2014 (post-reorganization). We selected these years based on the availability of Census data. The 2009 Census data were from the American Community Survey (ACS) 5-twelvemonth approximate, based on the data nerveless between 2005 and 2009. The ACS 2014 is the interpretation based on the survey betwixt 2010 and 2014. ACS uses the same measurement as the decennial Census but takes the average over a longer menstruation. Thus, the ACS data is more reliable than the single-year survey due to its larger sample size and temporal coverage.
Bird data were acquired from the USGS North American Breeding Bird Survey (BBS), a long-term and large-scale avian monitoring program. In this newspaper, we used the total number of species and the total bird population to summate a Shannon Index of Diversity (Equation 1) for each Bbs route from 2000 to 2015. Since there is no BBS road in the City of New Orleans, we selected the closest routes which are 33–136 km from the city heart.
where,
p i = proportion of the population made up of species i
south = number of species in sample
The land comprehend and land encompass change data were gathered from the Coastal Change Analysis Programme (C-CAP) of NOAA's Role of Coastal Management. The C-CAP regional country cover and change products are nationally standardized, raster-based inventories of land cover for the littoral areas, from the analysis of multiple dates of remote-sensing imagery with 30 meters/pixel resolution. The thematic country-embrace country-change raster files were input and processed using ArcGIS 10.3 to calculate area of change for the land type of interest given C-CAP'south 24 country-comprehend classes. Nosotros specifically examined the following information: (ane) developed expanse which is covered by physical, asphalt, and other synthetic materials; (2) vegetation that includes forest, scrub state, and grassland; and (3) wetlands (NOAA, 2017). Table ane shows the data, their sources, and temporal coverage available for this study. We used land-cover change over fourth dimension at the metropolis scale and across (southern Louisiana) as a surrogate of environmental degradation at different spatial scales. Zonal statistics were applied in ArcGIS to measure land-cover change and summate percent of change outside of the political boundary including open h2o.
Tabular array 1. Variables and information sources.
Level of heterogeneity or diverseness is a critical indicator of resilience. It reflects the options and a system's capacity to respond to change and disturbance in various ways (Walker and Salt, 2006). Inequality amid income and ethnic/racial groups has been an issue in New Orleans. In general, the lower-income population lives in the areas with higher risk to flooding (Kates et al., 2006). We used spatial autocorrelation to quantify the pattern of change and test the following hypotheses:
H1: The spatial distribution of some social and economic variables became less amassed (exhibited more heterogeneity) than pre-Katrina condition.
H2: Disproportional risk to flooding decreased over fourth dimension.
To test the start hypothesis, we used a spatial autocorrelation index, Moran's I (Equation ii), to measure out the correlation of targeted variables and determine its spatial pattern (cluster, random, or detached). Moran'south I values range between −1 and 1, indicating that attribute values at side by side geographic sites are more dissimilar (−one) or more similar (1).
For an observation Z at location i ( , where is the mean of variable 10): W ij is the chemical element of the spatial weights matrix, S 0 = ∑i∑j W ij is the sum of all the weights, and n is the number of observations. All the spatial autocorrelation analyses were performed in the Geoda i.viii.16.4 spatial data analysis software.
For the 2nd hypothesis, we mapped the expanse flooded during Hurricane Katrina in ArcGIS and superimposed temporal Census data on the GIS map to calculate the population of each ethnicity in the area over time. The flood-damaged information comes from a report (Logan, 2006), which used FEMA and high-resolution images from the Dartmouth Flood Observatory to estimate flood-impacted surface area right after Hurricane Katrina.
Lastly, we synthesized the social-ecological weather condition and monitored change over time using a deductive statistical approach, principal component analysis, to characterize this coastal system, and observe how the social-environmental organisation of New Orleans changed and restructured over time.
Assay and Results
Land-Cover Modify
An assessment of land-cover changes pre- and post-Katrina (2005–2006) showed that the primary land conversion in both southern Louisiana and New Orleans was from wetland loss. Deforestation and destruction of human being-made infrastructure was also disquisitional (Tables 2, iii). Over the longer term (2001 to 2010), southern Louisiana experienced deforestation reflected in more than vegetation changing from forest and grassland to shrubs (Table four). At a finer scale, the major land modify in New Orleans resulted from wetland loss and new development (Table 5).
Table ii. Land-cover change between 2005 and 2006 in Southern Louisiana.
Table 3. State-cover change betwixt 2005 and 2006 in New Orleans.
Table 4. Land-cover change between 2001 and 2010 in Southern Louisiana.
Table 5. Land-cover change between 2001 and 2010 in New Orleans.
Bird Diversity Change
We selected five Bulletin board system routes close to New Orleans. Three routes are in northeast, south, and southwest of Louisiana, and two routes are in neighboring Mississippi with the distances ranging from 33 to 136 km away from the city middle. A plot of the Shannon variety index displays the index value for each route from 2000 to 2015 and the mean for the 2 LA routes (A and B) closest to New Orleans (Figure 1). (Note that the discontinuities relate to missing information (2004 for route B; 2010 for route E; 2001–2004, 2016 for road C). At that place was a slight drop for these ii routes after 2006, in 2008, but both tended to vary around the xv-twelvemonth average level and by 2010, they decreased once more until 2014.
Figure 1. Bird variety represented by Shannon Alphabetize betwixt 2000 and 2015. LA-Louisiana; MS-Mississippi; Dot lines are 15-year average of the two sites in Louisiana.
Income Change
Income is an indicator associated with socio-economic condition and may serve equally a surrogate measure of a group'due south ability to cope with change. We calculated z-scores of incomes at Demography-tract level between 2000 and 2014 to monitor changes in wealth condition. The z-score is the standard deviation from the mean of all tracts for a specific fourth dimension step. A positive z-score meant the income level was higher up the mean (of 0) and a negative z-score meant the observation was below the average. Using z-score to examine the income distribution yielded a relative indication of how each tract's income status compared to the other tracts in New Orleans over the period. The data yielded four wealth categories. For example, in Figure 2:
1. Remained relatively low income: Neighborhoods with z-scores below 0 betwixt 2000 and 2014.
ii. Decreasing wealth: Neighborhoods with a z-score above 0 in 2000 and below 0 in 2014.
3. Remained wealthy: Neighborhoods with z-scores above 0 in both 2000 and 2014.
iv. Increasing wealth: Neighborhoods with a z-score beneath 0 in 2000 and above 0 in 2014.
Effigy 3 reveals the geographical location of these four types of neighborhoods with dissimilar fourth dimension steps: (1) between 1990 and 2000; and (2) between 2000 and 2014.
Figure 2. Z scores on distribution of income between 2000 and 2014.
Figure three. Spatial distribution of income change between (1) 1990 and 2000; and (two) 2000 and 2010.
Since some tract boundaries were modified past the US Census Bureau over time, we fabricated the census information comparable by aligning historical census information to year 2010 Census boundaries, using the Longitudinal Tract Data Base of operations plan developed by Brownish Academy. The program applies proportional surface area weighting to assign demography variable values to the consistent spatial unit (Logan et al., 2016). Figure iii maps the neighborhoods that remained wealthy, relatively depression income, and those with increasing or decreasing wealth in two-time segments: 1990–2000 and 2000–2014).
Spatial Autocorrelation of Income, Unemployment, Vacancy Rates, Renters and Owners
Spatial autocorrelation of income, unemployment, vacancy rates, renters, and owners. The spatial analysis reveals that while occupied housing (renter and owner) and vacancy rate dispersed (Moran's I declined), income became more than clustered over time (Moran's I increased), the unemployment charge per unit exhibited more than variability: declined from pre-tempest rates (2005–2009) and by 2014 spatial autocorrelation had increased (Figure 4). Specifically, the spatial analysis reveals that income patterns became more clustered over fourth dimension, meaning wealthy and low-income neighborhoods were spatially autocorrelated, instead of randomly distributed across the metropolis. Moreover, while high-income clusters grew slightly between 2005 and 2009, low-income clusters contracted (relatively concentrated), just both became more than aggregated, respectively, in 2014. Accordingly, the aggregation pattern measured by Moran's I suggests that income inequality enlarged spatially over fourth dimension.
Figure iv. Cluster maps and Moran'south Index of income, housing characteristics, unemployment, and vacancy. Areas in dark blue are the clusters of low value, and areas in red are the clusters of high value. The values relate to loftier or low income, unemployment rate, etc.
The average unemployment rate at Demography-tract scale in New Orleans was 10.96%, higher than the national level in 2000 (four.00%). Subsequently Hurricane Katrina, the average unemployment rate reached xiii.92%, and so dropped to 12.74% in 2014. Further, there was variation in unemployment charge per unit at a fine calibration inside the city. In 2014, almost 20% of neighborhoods had unemployment rates beneath v%, still nearly as many (16%) neighborhoods had unemployment rates in a higher place 20%. The caste of spatial autocorrelation decreased later on Hurricane Katrina simply increased once more in 2014. Hot spots of high unemployment were less aggregated, but a new cluster appeared on the e side of the metropolis in 2014.
The vacancy rate during 2005-09 was about 26.71%, more than twice the level in 2000 (12.68%). The charge per unit decreased to 21.29% in 2014, suggesting that fifty-fifty 9 years after Katrina, there are notwithstanding neighborhoods with limited chapters for reorganization. Meanwhile, the spatial patterns of vacant units become less aggregated over fourth dimension but remained spatially autocorrelated. As Figure four shows, the hot spot of vacant units clustered in the urban cadre. Specifically, the geographical location of highly vacant areas changed and became more discrete later on Katrina, yet the spatial cluster returned to a pattern like to pre-Katrina status by 2014.
The spatial distribution of renter and owner-occupied housing units was highly clustered. Though the degree of spatial autocorrelation has decreased since 2000, the hot spot of owner-occupied and renter-occupied units remained segregated.
Principal Component Analysis
Principal component analysis (PCA) is a multivariate statistical approach used to place the blueprint of similarity amid observations (Abdi and Williams, 2010). We input all variables from Tabular array 1, except for bird diversity data (considering information technology simply showed minor variation) and performed PCA in IBM SPSS 24. Varimax rotation was practical to the dataset.
Five master components were identified for the year 2000 with 73.14% of variance explained. Figure v shows the spatial distribution of the first three components. Component 1 accounts for 31.34% of the variance in the information. Results indicate a strong positive loading (values ≥ |0.50|) for median household income, median home value, median hire, non-Hispanic white, Hispanic, and Native Indian populations, and population living in a different state within the past 5 years. The component had negative loading for African Americans, unemployment charge per unit, and population living in the aforementioned house within 5 years. Component ii accounts for xx.46% of the variance with heavy loading on vacancy rate, unemployment rate, renter, and medium to high intensity of urbanization and negative loading on income, owner-occupied housing, and low adult areas. The third component explains 9.84% of the variance, with only two heavy and positive loadings for Asian population and wetland area. Component four accounts for half dozen.34% of the variance with strong positive loading for population living in different houses merely the aforementioned city inside five years. Lastly, Component v has 5.16% variance, with strong loading in income and vegetation comprehend (Table 6).
Effigy 5. First iii principal components for 2000. Classified past standard departure.
Table 6. Results of chief component analysis using data from year 2000.
The PCA after Hurricane Katrina reveals changes in the social-ecological system of New Orleans in 2009, with 71.64% of the variance explained through six components. Figure 6 shows the spatial distribution of the first iii components. Component 1 accounts for 26.55% of the variance, with strong positive loading in income, home value, rent, not-Hispanic white population and negative loading in percentage of African American population and unemployment charge per unit. Component 2 explains 15.10% of the variance and has heavy loading on renters, and medium to highly urbanized area forth with negative loading on possessor-occupied housing and low developed area. Component iii accounts for x.55% of the variance with heavy loading in Asian population and wetland surface area. Component four explains 7.56% of the variance and reflects strong positive loading in population living in the aforementioned house 1 yr before and negative loading in population living in a different house in the aforementioned city ane year before. Component 5 accounts for 5.99% of the variance with strong positive loading in Hispanic population and vacancy charge per unit. Lastly, Component 6 accounts for v.89%, and has heavy negative loading in newcomers and strong positive loading in population living in the same house i yr earlier (Table seven).
Figure 6. Commencement three principal components for 2005–2009. Classified by standard deviation.
Table 7. Results of master component analysis using data from year 2009.
In 2014, five components explained 74.05% of the variance. The spatial distribution of the first 3 components were mapped in Figure 7. Component one accounted for 29.6% of the variance, and a heavy positive loading in income, home value, hire, non-Hispanic white population, and strong negative loading in African American population, unemployment rate, and new comers. Component 2 accounts for 20.93% of the variance with strong positive loading in renter, population living in dissimilar house but the same city within 1 twelvemonth and negative loading in owner-occupied housing and population living in the same business firm 1 year before. Component three explains 11.52% of the variance, and has potent positive loading in vacant rate, renter, and highly urbanized expanse, and stiff negative loading in owner-occupied housing. Component iv accounts for six.14% of the variance and has heavy positive loading in percentage of Asian population and wetland area. Lastly, Component 5 that accounts for 5.87% of the variance, and has strong positive loading in percentage of Native Indian population and vegetation comprehend (Table 8).
Figure seven. Commencement three principal components for 2014. Classified by standard difference.
Tabular array 8. Results of principal component analysis using data from yr 2014.
Population Living in Flooded Areas
Figure eight shows the demographic modify in flooded areas over fourth dimension. A large proportion of African Americans live in the urban center and consequently, reside in flooded areas. The portion of African Americans living in the flooded areas decreased slightly after Hurricane Katrina, but they remain the largest population living in the flooded areas.
Figure viii. Demographic change in flooded areas over fourth dimension.
Discussion
Through investigating social, economical and environmental data temporally and spatially, we sought to better empathize how the human being and natural environment responded to Hurricane Katrina and how these systems reorganized and recovered from the devastating event. Hurricane Katrina decimated the Metropolis of New Orleans, and later several years of reconstruction, the metropolis is growing with some innovations. Liu and Plyer (2010, p. 6) claim the city "has go more resilient, with increased civic capacity and new systematic reforms, ameliorate positioning the metro area to adapt and transform its futurity" using certain economic indicators (e. k., increase in wage and income, growing entrepreneurship). Nevertheless, there was still an outstanding question as to whether all residents do good from this? Are the ecology concerns well-understood and managed? Accordingly, there was a need for a systematic and integrated arroyo to evaluate changes in the social and environmental condition of New Orleans.
Income status was more stagnant between 1990 and 2000 than between 2000 and 2014. During the initial period, near of the census tracts (96 out of 173 tracts; 55.50%) were below the metropolis's average, 64 tracts were relatively stable and wealthy and footling alter was evident given that just eight neighborhoods were considered declining and 5 tracts were growing in income. Between 2000 and 2014 (encompassing the effects of Hurricane Katrina) income status became more dynamic. There were eighteen neighborhoods that experienced decreasing income, 20 tracts with growing income, 48.5% (84 tracts) were beneath the city's boilerplate and the remaining 51 tracts were in stable wealthy status. From a paired sample t-examination, the income status was significantly different later Hurricane Katrina than before (p <0.05). The increment in average income can exist viewed as a sign of growth. Nevertheless, when looking at the spatial distribution of neighborhoods that remained beneath the urban center's average, inner metropolis residents were not meliorate off later the system reorganized.
Carpenter and Brock (2008) use the term "poverty trap" as a metaphor to describe a social-ecological system's adaptive capacity. In a social-ecological poverty trap, the arrangement has low or loose connexion and resilience. The potential for change is non realized because the system lacks resources to reorganize and movement forward (Gunderson and Holling, 2002; Westley, 2006; Carpenter and Brock, 2008). The situation in New Orleans before 2000 and before Hurricane Katrina could be described every bit in a "poverty trap," such that while more than than one-half of the neighborhoods were in a relatively low-income state of affairs and only v Census tracts had increased income betwixt 1990 and 2000. In the aftermath of Katrina, the dynamics of income status changed. From 2000 to 2010, at that place were twenty neighborhoods experiencing increased income equally the boilerplate income increased in the post-Katrina era.
While bird diversity was relatively stable, land-utilize patterns along with economic growth in New Orleans put pressure on natural systems and potentially damaged the area's long-term sustainability (Gotham et al., 2014). Afterwards Hurricane Katrina, these ecology concerns remain and potentially undermine the social-ecological resilience of New Orleans to natural disasters. These concerns include the continuous loss of wetlands at both the city and southern Louisiana scale, high intensity of urbanization inside the metropolis without addressing tempest water and flooding bug, and loss of forests in southern Louisiana.
Principal Component Analysis Pre- and Post-katrina
Pre-Katrina Weather (2000)
In 2000, social-ecological conditions were diverse (Figure 5). Wealthy populations co-existed with not-rich urban dwellers in the eye of the city. Component 1 which accounts for 31.34% of variance of the data revealed that the groups/areas with high-income were non-Hispanic whites with high-holding and rental value and independent a loftier percentage of newcomers (population who moved from other states at least 1 year prior to the survey). The areas with very high component one scores (> two.5 standard deviation) were Key Business District, Lower Garden Commune, French Quarter, Marigny, Audubon, and New Aurora. Some of these neighborhoods (east.g., French Quarter), besides contained pockets of low-income renters, living in the highly urbanized places that are positively associated with unemployment and vacancy rates. Regarding ecological characteristics, we found that income-level was positively associated with high vegetation encompass. This relationship is consistent with social stratification theory, which presumes high socioeconomic status population is very likely to take more (or ameliorate access to) environmental amenities including dark-green spaces (Grove et al., 2006; Roy Chowdhury et al., 2011).
Postal service-Katrina Conditions (2005–2009)
After Hurricane Katrina, the outset principal that explained the well-nigh variance beyond the metropolis was primarily linked to wealthy Caucasians. Newcomers were no longer associated with Component 1 and the geography and variability in Component i changed (Effigy half dozen). Central Business District, Lower Garden District, and French Quarter no longer received the highest scores (but their scores were still relatively loftier, compared with the residue of the city). Nosotros did not find the highest scores in the metropolis heart only instead, the neighborhoods with the highest Component ane scores were found in Audubon, followed past Lakeshore, Lake View, Terrace & Oaks, and New Aurora. Component 2 accounts for 15% of the variance and is characterized by renters with relatively depression income in highly developed areas and in the aftermath of Katrina, loftier vacancy and unemployment rates were no longer fundamental characteristics under Component two. With regard to the spatial distribution of Component 2, the neighborhoods with highest scores were withal found in the city center, suggesting that low-income and lack of buying of housing in highly urbanized areas remains an inner-city trouble. In the post-Katrina era, vacancy charge per unit became positively associated with Hispanic population (Component 5).
More Recent Conditions (2010–2014)
Afterward several years of recovery, the characteristics of Component 1 (wealthy Non-Hispanic White population, high percentage of newcomers, high dwelling values and rent) became like to pre-disaster weather. The loftier scores of Component 1 were observed in Lakewood, Lakeshore, Audubon, Central Business concern District, and French Quarter (Figure seven). Relatively low-income renters were divided into two components—i grouping (Component 2) with a positive human relationship with the population who did not live in the same business firm in the city one year prior to the survey, and the other (Component iii) holds positive relationships with vacancy charge per unit and percentage of highly developed area. Interestingly, vegetation cover was no longer associated with income and instead was linked with native Indian population in 2014. Studies accept shown that land abandonment or unmanaged vacant lots are the driver of emergent vegetation in New Orleans (Lewis et al., 2017) and in a shrinking metropolis context (Schwarz et al., 2018).
Green infrastructure (natural and man made) is considered an important ecosystem service as well as an ecology amenity in many environmental studies (Greenish et al., 2016). Typically, quantitatively assessing the level of greenness and identifying the spatial unevenness of vegetation cover are the major approaches to evaluating provision of ecosystem services and equality. In the city of New Orleans, the amount of vegetation comprehend was one time positively associated with household income. However, after the system reorganized, maintaining vegetation in good condition turned into a claiming in minority, low-income neighborhoods which accept abandoned land. Within this context, evaluating the amount of vegetation does not provide enough information to characterize ecosystem service provision and the impact on environmental justice (Lewis et al., 2017). Moreover, the assessment of equity, which is determined by the quality of outcomes, will be more critical to improve understand ecosystem services in the postal service-disaster era.
Conclusion
Natural hazards tin can trigger collapse merely also create opportunities for systems to learn, restructure, and reorganize to manage disaster resilience. Our study examines the social and ecological condition of New Orleans (and surrounding areas) before and subsequently Hurricane Katrina. By analyzing the modify in arrangement status using social, economic and environmental factors, we identified some of the characteristics of the system's regrowth and reorganization trajectories. Although the ongoing population recovery may be a sign of revitalization, the metropolis and metropolitan area go on to confront socioeconomic inequalities and vulnerability to natural disasters. Our findings propose that high poverty rates in some areas, and environmental concerns such every bit loss of bird variety and wetlands, create challenges to the sustainability of the city. The spatial distribution of social-ecological conditions over time reveals certain levels of change and reorganization after Katrina, just the reorganization did non interpret into greater equity.
Resilience is not static. Assessing disaster resilience requires the measurement of changing atmospheric condition and the reorganization process. Our analyses comprised three-time steps, including earlier and later the organisation was disturbed, and demonstrate an advanced approach for assessing disaster resilience. In addition to temporal aspects, we examined spatial dimensions of disaster resilience to include capturing patterns in social, economic and environmental conditions. We advise that as loftier-resolution time-series data becomes bachelor, future research should include monitoring long-term spatial heterogeneity of other environmental variables (east.g., terrestrial biodiversity, land use zoning, green infinite, vacant lots, and overflowing depth).
Disaster resilience cannot be fully understood by collecting but one-time-step data or using data that insufficiently capture the area and menses of study. Accordingly, in addition to larger and finer temporal coverage, the spatial resolution is important. For example, to accost the inter-scalar interactions, it is critical to have data at both local and regional scales. In our analysis, Breeding Bird Survey data has excellent temporal resolution, however, its spatial resolution is limited. For cess of disaster resilience, it would be useful to have ecological variables from a very fine scale (such as high-resolution Lidar data) with continuous temporal coverage. Regarding the social and economic variables, earlier 2009, the Census data was only bachelor every 10 years, which reduces a researcher's ability to mensurate dynamics of socio-economic alter. More recently, the American Community Survey began offering continuous information collections with finer temporal resolution and multiple spatial scales; hence, we were able to incorporate this information. Although the lack of bachelor data limited our power to perform a more consummate ecosystem cess, this work provides central advances in inquiry for disaster resilience. In particular, introducing the dynamic, time-stride analysis, employing a social-ecological resilience lens and incorporating ecological variables (where available) were significant improvements for research in disaster resilience. Lastly, we highlight the need to not merely examine system conditions quantitatively, but also qualitatively (e.g., governance, or quality of the green space/dark-green infrastructure) to link with management options.
While, this research mainly focuses on Hurricane Katrina's bear upon on New Orleans, our assessment of human and natural systems is not limited to the municipal scale and provides a modular framework for assessing impacts at multiple geographic scales (i.e., other variables may exist added and it is possible to expand the scale and scope). Man and natural activities may occur at a detail location but impacts are non constrained past municipal boundaries; hence, understanding the cross-scale implications of social-ecological alter is critically important for disaster resilience (Green et al., 2015).
Author Contributions
W-CC conceived the original idea, designed the research, and performed the analyses. Due west-CC, TE, CR, and AG developed the manuscript. CR gathered and processed the Breeding Bird Survey Information from USGS.
Conflict of Interest Statement
The authors declare that the inquiry was conducted in the absence of whatever commercial or financial relationships that could be construed equally a potential disharmonize of interest.
Acknowledgments
The findings and conclusions in this manuscript accept non been formally disseminated by the U.S. Environmental Protection Agency and should not exist construed to represent any bureau decision or policy. We thank the two reviewers for their constructive feedback that helped improve this manuscript.
References
Abdi, H., and Williams, Fifty. J. (2010). Principal component analysis. Wiley Interdisc. Rev. Comput. Stat. ii, 433–459. doi: 10.1002/wics.101
CrossRef Total Text | Google Scholar
Allen, C. R., Angeler, D. G., Garmestani, A. S., Gunderson, L. H., and Holling, C. S. (2014). Panarchy: theory and application. Ecosystems 17, 578–589. doi: 10.1007/s10021-013-9744-2
CrossRef Total Text | Google Scholar
Angeler, D. 1000., Allen, C. R., Garmestani, A. Due south., Gunderson, L. H., and Linkov, I. (2016). Panarchy utilise in environmental science for gamble and resilience planning. Environ. Syst. Decis. 36, 225–228. doi: 10.1007/s10669-016-9605-vi
CrossRef Full Text | Google Scholar
Bruneau, M., Chang, S. E., Eguchi, R. T., Lee, Yard. C., O'Rourke, T. D., Reinhorn, A. M., et al. (2003). A framework to quantitatively assess and enhance the seismic resilience of communities. Earthquake Spectra 19, 733–752. doi: ten.1193/ane.1623497
CrossRef Full Text | Google Scholar
Carpenter, S., Walker, B., Anderies, J. Yard., and Abel, N. (2001). From metaphor to measurement: resilience of what to what? Ecosystems 4, 765–781. doi: 10.1007/s10021-001-0045-ix
CrossRef Full Text | Google Scholar
Carter, N. (2005). New Orleans, Levees, and Floodwalls: Hurricane Damage Protection. Washington, DC: Congressional Enquiry, Service.
Google Scholar
Chuang, W. C., Garmestani, A., Eason, T. N., Spanbauer, T. L., Fried-Petersen, H. B., Roberts, C. P., et al. (2018). Enhancing quantitative approaches for assessing community resilience. J. Environ. Manage. 213, 353–362. doi: 10.1016/j.jenvman.2018.01.083
PubMed Abstract | CrossRef Full Text | Google Scholar
Cutter, S. L., Ash, 1000. D., and Emrich, C. T. (2014). The geographies of customs disaster resilience. Global Environ. Change 29, 65–77. doi: 10.1016/j.gloenvcha.2014.08.005
CrossRef Total Text | Google Scholar
Cutter, S. Fifty., Barnes, 50., Berry, Thou., Burton, C., Evans, E., Tate, E., et al. (2008). A place-based model for agreement community resilience to natural disasters. Global Environ. Modify 18, 598–606. doi: x.1016/j.gloenvcha.2008.07.013
CrossRef Full Text | Google Scholar
Cutter, S. L., Boruff, B. J., and Shirley, Due west. L. (2003). Social vulnerability to environmental hazards. Soc. Sci. Quart. 84, 242–261. doi: 10.1111/1540-6237.8402002
CrossRef Full Text | Google Scholar
Cutter, S. L., and Finch, C. (2008). Temporal and spatial changes in social vulnerability to natural hazards. Proc. Natl. Acad. Sci. U.South.A. 105, 2301–2306. doi: x.1073/pnas.0710375105
PubMed Abstract | CrossRef Total Text | Google Scholar
Dixon, T. H., Amelung, F., Ferretti, A., Novali, F., Rocca, F., Dokka, R., et al. (2006). Subsidence and flooding in New Orleans. Nature 441:587. doi: 10.1038/441587a
PubMed Abstract | CrossRef Full Text | Google Scholar
Eakin, H., Winkels, A., and Sendzimir, J. (2009). Nested vulnerability: exploring cross-scale linkages and vulnerability teleconnections in Mexican and Vietnamese java systems. Environ. Sci. Policy 12, 398–412. doi: 10.1016/j.envsci.2008.09.003
CrossRef Total Text | Google Scholar
Elliott, J. R., and Pais, J. (2006). Race, class, and Hurricane Katrina: social differences in man responses to disaster. Soc. Sci. Res. 35, 295–321. doi: 10.1016/j.ssresearch.2006.02.003
CrossRef Full Text | Google Scholar
Fekete, A. (2009). Validation of a social vulnerability index in context to river-floods in Germany. Nat. Hazards World Syst. Sci. 92, 393–403. doi: x.5194/nhess-9-393-2009
CrossRef Total Text | Google Scholar
Folke, C. (2006). Resilience: the emergence of a perspective for social–ecological systems analyses. Global Environ. Change sixteen, 253–267. doi: 10.1016/j.gloenvcha.2006.04.002
CrossRef Full Text | Google Scholar
Gallopín, G. C. (2006). Linkages between vulnerability, resilience, and adaptive capacity. Global Environ. Change 16:293–303. doi: ten.1016/j.gloenvcha.2006.02.004
CrossRef Full Text | Google Scholar
Goldenberg, Due south. B., Landsea, C. Due west., Mestas-Nuñez, A. M., and Grey, West. M. (2001). The recent increase in Atlantic hurricane activity: causes and implications. Science 293, 474–479. doi: ten.1126/science.1060040
PubMed Abstract | CrossRef Total Text | Google Scholar
Gotham, Chiliad., and Campanella, R. (2011). Coupled vulnerability and resilience: the dynamics of cantankerous-calibration interactions in postal service-Katrina New Orleans. Ecol. Soc. xvi:ane. doi: 10.5751/ES-04292-160312
CrossRef Full Text | Google Scholar
Gotham, K. F., Blum, M., and Campanella, R. (2014). Toward a new normal: trauma, diversity, and the New Orleans Urban Long-Term Research Area Exploratory (ULTRA-Ex) Project. Cities the Environ. seven:4.
Google Scholar
Dark-green, O.O., Garmestani, A. S., Allen, C. R., Gunderson, Fifty. H., Ruhl, J. B., Arnold, C. A., Chaffin, B. C., et al. (2015). Barriers and bridges to the integration of social-ecological resilience and law. Front end. Ecol. Environ. 13, 332–337. doi: ten.1890/140294
CrossRef Full Text | Google Scholar
Green, O. O., Garmestani, A. Due south., Albro, S., Ban, North. C., Berland, A., Burkman, C. East., et al. (2016). Adaptive governance to promote ecosystem services in urban green spaces. Urban Ecosyst. 19, 77–93. doi: 10.1007/s11252-015-0476-2
CrossRef Full Text | Google Scholar
Grove, J. M., Cadenasso, M. L., Burch, W. R., Pickett, South. T. A., Schwarz, K., O'Neil-Dunne, J., et al. (2006). Data and methods comparing social Structure and vegetation structure of urban neighborhoods in Baltimore, Maryland. Soci. Nat. Resour. 19, 117–136. doi: ten.1080/08941920500394501
CrossRef Full Text | Google Scholar
Gunderson, L. H., and Holling, C. S. (2002). Panarchy : Understanding Transformations in Human being and Natural Systems. Washington, DC: Isle Press.
Google Scholar
Holling, C. S. (1973). Resilience and stability of ecological systems. Ann. Rev. Ecol. Systemat. iv, 1–23. doi: 10.1146/annurev.es.04.110173.000245
CrossRef Full Text | Google Scholar
Kates, R. Due west., Colten, C. E., Laska, S., and Leatherman, S. P. (2006). Reconstruction of New Orleans after Hurricane Katrina: a enquiry perspective. Proc. Natl. Acad. Sci. The statesA. 103, 14653–14660. doi: 10.1073/pnas.0605726103
PubMed Abstract | CrossRef Full Text | Google Scholar
Kelly, P. M., and Adger, W. Due north. (2000). Theory and practise in assessing vulnerability to climate change and facilitating adaptation. Clim. Modify 47, 325–352. doi: ten.1023/A:1005627828199
CrossRef Total Text | Google Scholar
Lewis, J. A., Zipperer, W. C., Ernstson, H., Bernik, B., Hazen, R., Elmqvist, T., et al. (2017). Socioecological disparities in New Orleans post-obit Hurricane Katrina. Ecosphere eight, 1–24. doi: 10.1002/ecs2.1922
CrossRef Full Text | Google Scholar
Liu, A., and Plyer, A. (2010). An Overview of Greater New Orleans: From Recovery to Transformation. New Orleans Index at Five. Washington, DC: Brookings Establishment and Greater New Orleans Community Information Center.
Google Scholar
Logan, J. (2006). The Impact of Katrina: Race and Class in Storm-Damaged Neighborhoods: Spatial Structures in the Social Science, Providence: Brown University.
Google Scholar
Logan, J. R., Issar, Due south., and Xu, Z. (2016). Trapped in Place? Segmented resilience to hurricanes in the Gulf coast, 1970-2005. Census 53, 1511–1534. doi: 10.1007/s13524-016-0496-iv
PubMed Abstruse | CrossRef Full Text | Google Scholar
Norris, F. H., Stevens, Due south. P., Pfefferbaum, B., Wyche, G. F., and Pfefferbaum, R. L. (2008). Community resilience as a metaphor, theory, set of capacities, and strategy for disaster readiness. Am. J. Community Psychol. 411–412, 127–150. doi: x.1007/s10464-007-9156-6
CrossRef Full Text | Google Scholar
Opdyke, A., Javernick-Volition, A., and Koschmann, 1000. (2017). Infrastructure hazard resilience trends: an analysis of 25 years of research. Nat. Hazards 872, 773–789. doi: 10.1007/s11069-017-2792-8
CrossRef Full Text | Google Scholar
Roy Chowdhury, R., Larson, K., Grove, Grand., Polsky, C., Melt, East., Onsted, J., et al. (2011). A Multi-scalar arroyo to theorizing socio-ecological dynamics of urban residential landscapes. Cities Environ. four, six–nineteen. doi: x.15365/cate.4162011
CrossRef Total Text | Google Scholar
Schwarz, One thousand., Berland, A., and Herrmann, D. L. (2018). Green, merely not but? Rethinking environmental justice indicators in shrinking cities. Sustain. Cities Soc. 41, 816–821. doi: 10.1016/j.scs.2018.06.026
CrossRef Full Text | Google Scholar
Sikula, Northward. R., Mancillas, J. W., Linkov, I., and McDonagh, J. A. (2015). Take a chance management is not enough: a conceptual model for resilience and adaptation-based vulnerability assessments. Environ. Syst. Decisions 35, 219–228. doi: 10.1007/s10669-015-9552-7
CrossRef Full Text | Google Scholar
Summers, J. K., Smith, L. M., Harwell, Fifty. C., and Buck, K. D. (2017). Conceptualizing holistic community resilience to climate events: foundation for a climate resilience screening index. GeoHealth 1, 151–164. doi: 10.1002/2016GH000047
PubMed Abstruse | CrossRef Full Text | Google Scholar
Timmerman, P. (1981). Vulnerability, Resilience and the Collapse of Social club. Toronto, ON: Institute for Ecology Studies, University of Toronto.
Google Scholar
Turner II, B. 50., Kasperson, R. East., Matson, P. A., McCarthy, J. J., Corell, R. Westward., Christensen, L., et al. (2003). A framework for vulnerability assay in sustainability science. Proc. Natl. Acad. Sci. U.S.A. 100:8074. doi: 10.1073/pnas.1231335100
CrossRef Total Text | Google Scholar
van der Leeuw, S. (2013). "Resilience and patterns of alter," in Sustainable World, ed. S. Remington-Doucette (Dubuque, IA: Kendall Chase).
Google Scholar
Walker, B., and Salt, D. (2006). Resilience Thinking: Sustaining Ecosystems and People in a Changing World. Washington, DC: Washington: Island Press.
Google Scholar
Webster, P. J., Holland, G. J., Back-scratch, J. A., and Chang, H. (2005). Changes in tropical cyclone number, duration, and intensity in a warming environment. Science 309, 1844–1846. doi: x.1126/scientific discipline.1116448
PubMed Abstruse | CrossRef Full Text | Google Scholar
Westley, F. (2006). Getting to Possibly : How the World is Inverse. Toronto ON: Random House Canada.
Google Scholar
Wisner, B. (2004). "Assessment of capability and vulnerability," in Mapping Vulnerability : Disasters, Development, and People, eds. G. Bankoff, M. Frerks, and D. Hilhorst (London: Earthscan).
Google Scholar
Globe Bank (2013). Building Resilience: Integrating Climate and Disaster Take a chance into Development. Washington, DC: Globe Bank.
Google Scholar
jeffersonpand1948.blogspot.com
Source: https://www.frontiersin.org/articles/10.3389/fenvs.2019.00068/full
0 Response to "Peer Reviewed Papers on Hurricane Katrina Environmental Impacts"
Postar um comentário