Hazard adaptation in urban New Orleans
and Houston: comparing societal
responses to hurricanes
Extreme weather events affect society in multiple ways: political, economical, ecological, and social. Such complex natural stressors have interrelated elements that influence regions in diverse manners and ought to be studied in a comprehensive and interdisciplinary fashion. The aim of this paper is to evaluate the process of learning from past adaptation methods, while identifying new vulnerabilities. The main question answered is: how do past adaptive measures to hurricane Katrina in New Orleans compare to current adaptive measures regarding hurricane Harvey in Houston from an interdisciplinary perspective? A literature study is performed on the cases of hurricane Katrina in New Orleans and hurricane Harvey in Houston.
Keywords: Harvey, Katrina, Houston, New Orleans, adaptation, natural hazards, complexity.
Students: Koen Vermeij 11032340 Max van der Boog 11002719 Joris den Breejen 10716637 Ingmar Meeboer 10274421 Tutor: Fenna Hoefsloot
Expert: Kenneth Rijsdijk Course: Interdisciplinary Project Assignment: Final Paper
Table of contents
Introduction. ...3
Interdisciplinary Integration...4
Theoretical Framework...4
Problem definition and complexity... 11
Selected method and data...12
Spatial Analysis Method... 12
Results – New Orleans...12
Results - Houston...18
Results - Comparison...23
Conclusion & Discussion...25
References...26
In 2005, a record-breaking hurricane hit the US city of New Orleans, killing over 1800 individuals, and causing over $100 billion worth of damage (NHC-NOAA-NWS, 2011). More recently, Houston was hit by hurricane Harvey, which killed 82 people (The Washington Post, 2017). At the end of 2017, the damage is estimated to be over $180 billion, making it the most costly natural disaster to have ever hit the US (Amadeo, 2017).
New Orleans and Houston share some geographical, political and demographic settings. They are both located on the shores of the Gulf of Mexico, positioned along the same latitude, share the same federal jurisdictional system, and have relatively low population densities. This data and some more aspects related to the setting of both cities is displayed in table 1. Such similarities make it viable to study how both cities responded to a comparable natural hazard. Both cases will be analysed on how social, economic, environmental and political interrelated factors regarding the adaptability of the city functioned and behaved before and after the impact of the hurricanes. Through this comparative assay this study attempts to evaluate if mistakes made previously in hazard adaptation were not repetitive, exploring elements from hurricane Katrina and Harvey respectively. Such analysis could identify where the potential for future hazard adaptation lies. Therefore, the main research question that this paper will attempt to answer is: how does past adaptation to hurricane Katrina in New Orleans compare to current adaptation regarding hurricane Harvey in Houston from an interdisciplinary perspective?
Hurricanes are expected to intensify and appear more frequently in the future as a result of global environmental change (Pachauri et al., 2014). Identifying weaknesses in urban adaptation could help future efforts to improve protection against extreme weather events. Moreover, this study contributes to the existing scientific literature by being one of the first to evaluate the disaster in Houston by comparing it with the case of hurricane Katrina. The use of a interdisciplinary method is a requisite, as natural disasters, such as Katrina and Harvey, are immensely complex issues, spanning and interrelating multiple research and societal areas. Knowledge gained from a disciplinary perspective would not honor the complexity of these cases. The four disciplines that cooperate in this work are human geography, business administration, earth sciences and politics.
Firstly, the integration techniques used in this paper are explained. Subsequently, the theoretical framework is discussed, which is followed by the methodology. Then, the theories are used to analyse both cases in the results-section. Furthermore, the results are compared from which a conclusion is drawn.
New Orleans Houston
Region Gulf of Mexico Gulf of Mexico
Population 455.188 (year = 2005) 2.303.482 (estimate, year = 2016)
Size (km2) 906 km2 1.556 km2
Population density 892/km2 1.414/km2
Latitude 29.57 N 29.45 N
Longitude 90.4 W 95.22 W
Altitude -2m - + 6m + 32m
Damage ~ $100 billion ~ $180 billion
Fatalities 1833 82
Hurricane & category Katrina (cat. 3) Harvey (cat. 4) Table 1: Characteristics New Orleans & Houston (StatisticalAtlas, 2015)
Interdisciplinary Integration
In this paper, interdisciplinary is defined as being “a process in which members of different fields work together . . . to develop novel conceptual and methodologic frameworks with the potential to produce transcendent theoretical approaches” (Klein, 2008, p. 117). As stated earlier, the aim of this study is to transcend monodisciplinary perspectives by exposing overarching linkages between meta-level social conditions and more specific geophysical data. To connect these elements the integrative technique defined as the ‘organisation’ method is implemented; multiple academic domains combine and influence each other under an umbrella of organisational behaviour related to accountability frameworks explained in more detail in the theoretical framework (Rammelt, 2017). The organisation approach identifies similarities between concepts, redefines them, and organises, arranges or maps the causal links between them (Rammelt, 2017). A conceptual model is then constituted to integrate the differing perceptions of urban adaptability underlying earth sciences, politics, business
administration, and sociology (Menken & Keestra, 2016). Such relating concepts are naturally positioned within the theoretical framework, though they retain strong connections throughout the methods- and results-section as well.
Theoretical Framework
The foundation of the theoretical framework consists of theories derived from or applicable to the disaster that occurred in New Orleans as a result of hurricane Katrina. This case has been thoroughly researched on multiple aspects, contrary to the case of Houston and hurricane Harvey which lacks scientific examination. Therefore, the theories derived from the New Orleans case will perform as a theoretical basis for the analysis of both Katrina and Harvey. They are supplemented and critically discussed by theories from the related fields of research. The various disciplines from which these theories come, are earth sciences, political science, business administration, and human geography.
Through the integration of various theories and concepts a single framework emerges, which is visualised in Figure 1. The central theme within this framework is adaptation to natural hazards. This study defines adaptation in the context of human dimensions of global change, which usually refers to “a process, action or outcome in a system (household, community, group, sector, region, country) in order for the system to better cope with, manage or adjust to some changing condition, stress, hazard, risk or opportunity” (Smit & Wandel, 2006, p. 282). This broad definition of adaptation includes various actors, multiple scales, and both technical and social change. Moreover, adaptations can be anticipatory or reactive and autonomous or planned (Fankhauser et al., 1999; Smit et al., 2000). The following paragraphs comprise theories that describe, or are related to, adaptations to hurricane Katrina and the disaster caused by it.
Smith (2013) emphasizes anticipatory adaptation in advance of a natural hazard, which could contribute to reducing disaster vulnerability by preparing society for hazards through varying methods. The concept of vulnerability is described as the degree to which areas are expected to experience harm due to hazard exposure (Turner et al., 2003). The adaptation methods Smith mentions include preparedness, forecasts and warning systems, and land use planning. Preparedness is defined as the proactive planning – and testing – of hazard reduction actions on all time scales, accompanied by learning and training to identify risks and consider options (Smith, 2013; Chakraborty, et al., 2005). Preparedness therefore forms an overarching concept that includes most anticipatory adaptation measures. With risk is meant the chance that a potential danger results in an actual incident. Regularly, in assessment of risk the severity of the possible damage is taken into account as well. Chakraborty et al. (2005) state that early-warning systems and emergency evacuation improves local resilience, and they point at successful hurricane evacuation in urbanized coastal areas. Resilience refers to “the ability of a (social-ecological) system to tolerate disturbances and recover in such a way that essential features of the system return to the earlier state” (Baker & Refsgaard, 2007, p. 332). The idea of land use planning can be seen as an adaptation practice that has the purpose to use macro- and microzonation to control hazardous areas in a way that existing communities can be better protected, and any new development can be steered away from dangerous sites (Smith, 2013). Limitations in land use planning include a lack of knowledge, high costs of hazard mapping and the high difficulty in raising community awareness of risk (Godschalk et al., 1998). Burby and Dalton (1994) emphasize that hazard-based land use planning is likely to succeed with proper support of governments and local communities.
When discussing anticipatory and reactionary forms of adaptation it is useful to consider the concept of governance networks. This is because these networks comprehend all actors that are involved with the implementation of adaptation measures. According to Koliba, Mills and Zia, governance networks are:
Relatively stable patterns of coordinated action and resource exchanges involving policy actors crossing different social scales, drawn from the public, private, or nonprofit sectors and across geographic levels, who interact through a variety of competitive, command and control, cooperative, and negotiated arrangements for purposes anchored in one or more facets of the policy stream (2011, p. 211).
The characteristics of a governance network are interdependency, actors’ worldviews, complexity, institutionalisation and network management (Klijn & Koppenjan, 2012). The interactions and mutual reliance between actors form and sustain networks, while their strategies are based on their individual fundamental cognitive orientation and knowledge. This leads to complex dynamics within the network and the process of forming patterns of social relations and patterns of rules. As a result, there is a need for guidance and management of the interactions (Klijn & Koppenjan, 2012). Following from the description above, governance networks encompass the relevant actors and processes for the implementation of both anticipatory and reactionary adaptation measures. The processes of preparedness, forecasts and warning systems, and land use planning are thus incorporated in governance networks.
Koliba, Mills and Zia (2011) argue that good functioning accountability regimes are fundamental for effective performance of governance networks. They have created a framework in which three distinctive accountability frames and their basic elements of accountability are outlined (Table 2). Each frame describes the power relations between actors and demonstrates to whom account is rendered, the strength of these relations and what explicit standards or implicit norms constitute these relations. The ‘democratic frame’ involves the actors that can hold public administrations and their components accountable, and the institutions through which this is done. The ‘market frame’ includes actors that can hold accountable the market-oriented businesses, corporations, or firms that are involved in network activity, and the corresponding institutions that enable them. The ‘administrative frame’ “focuses on the processes, procedures, and practices that are employed in the administration and management of formally organized social networks” and “the dynamics operating between principals and agents, professionals, and collaborators” (p. 213). Within governance networks these accountability structures of various actors commingle, compete, or complement each other (Koliba, Mills & Zia, 2011). This results in various trade-offs between them, which can lead to indistinctness about which relations outdo others and undermine effective performance and adaptation measures.
Traditional liberal democratic theory argues that governance networks are hurting democracy because various responsible non-governmental actors in a network cannot be held accountable for
their performance through democratic institutions and are therefore undermining them. However, Sørenson and Torfing (2005) instead argue that the undermining of democracy is not necessarily the case. The extent to which democratic actors like representatives are engaged in setting up and governing these networks, and to which citizens can access these networks and profit from their benefits, are determining for the anchorage of democracy within networks and constitute ways to enhance it. Another critique can be pointed at the consumer accountability. In case of a disaster, the capability of consumers to hold corporations accountable diminishes as the possibility of choice in such a situation is being replaced by the necessity for supplies and the monopolies related to response and recovery efforts. Furthermore, limiting the market frame to consumer and shareholder accountability limits the space for other forms of accountability mechanisms that are concerned with the performance of business. For example, labor unions that exert influence on corporations, but also the idea of corporate social responsibility (CSR). CSR is a modern concept that creates certain expectations, both ethical and instrumental, for corporations. Therefore, it makes them accountable for deviating behaviour (Fox, Ward & Howard, 2002). According to Michael (2003), instrumental needs can result in improvements of the company’s goals, market share, profitability, and brand loyalty, making it a relevant contemporary corporate strategy. Following such instrumental incentives could misappropriate actual societal needs for profit: businesses deem specific CSR-related measures as socially desirable, but in-depth analysis evinces predominantly neo-liberal behaviour (Johnson et al., 2011). Instrumental CSR-related behaviour is not in the best interest of society as a whole and emphasizes corporate above local needs, obscuring the relations of accountability structures and invigorating potential for deviating behaviour. However, it could oppose the image consumers have of a firm. Mishra (2006) states that when a firm is serious, its publics perceive their communications as behavioral whereas superficial and inauthentic CSR efforts are more easily coined as purely symbolic. Lastly, following the argument of Sørenson and Torfing (2005), in case of a strong engagement of representatives and far reaching access for citizens within networks, market actors might also be held accountable indirectly through the mechanisms of the democratic frame. For example, in case of close engagement and tight control of democratic representatives over market actors within a network, citizens can hold market actors accountable through their representatives.
ACCOUNTABILITY FRAME ACCOUNTABILITY TYPE TO WHOM IS ACCOUNT RENDERED? STRENGTHOF TIES
EXPLICIT STANDARDS IMPLICIT NORMS DEMOCRATIC Elected
representative
Elected officials Strong (weaker when “lame duck”) Laws, statutes, regulations Representation of collective interests, policy goals Citizen Citizens Weak (stronger
during elections) Maximum feasible participation, sunshine laws, Deliberation, consensus, majority rule
deliberative forums
Legal Courts Strong Laws, statutes,
contracts
Precedence, reasonableness, due process, substantive rights
MARKET Shareholder/owner Shareholders/owners Strong Profit, performance measures
Efficiency
Consumer Consumers Weak Consumer law, product performance measures
Affordability, quality, satisfaction
ADMINISTRATIVE Bureaucratic Principals, supervisors, bosses Strong Performance measures, administrative procedures, organizational charts Deference to positional authority; unity of command, span of control
Professional Experts, professionals Weak (stronger when capacity to revoke licenses exists) Codes of ethics, licensure, performance standards Professional norms, expertise, competence
Collaborative Collaborators, peers, partners
Weak Written agreements, decision-making procedures, negotiation regimes Trust, reciprocity, durability of relationships Table 2: Accountability frames and structures (Koliba, Mills & Zia, 2011)
Closely related to the framework above is the notion of crisis management. Where the accountability framework focuses on the relationships between actors and to whom account must be rendered, crisis management concentrates more on the actual processes the network goes through in the situation of a disaster and is therefore primarily linked with reactionary adaptation. Together both theories contribute to the understanding of the dynamics of the governance network related to New Orleans. According to Comfort (2007), crisis management needs to facilitate four crucial processes: cognition, communication, coordination and control. Cognition is the ability to identify risk and act upon it. This is a function generally assigned to government agencies within a network. The earlier described concept of preparedness forms an important link with cognition, as there is correspondence with the aspects of risk identification and options consideration. Therefore, the pursuits before the striking of a hazard related to preparedness are crucial, as they generate the input for the good functioning of crisis management processes. Communication focuses on the capacity to create a shared understanding of the situation among crucial actors. Coordination comprehends aligning actors’ actions towards a common goal, while control comes down to keeping those actions aligned. Within this framework governmental actors have a central role in each of the four processes. Dualities in corporate incentives regarding adaptive measures can have negative effects on coordination efforts. Research by Johnson et al. (2011) suggests that the corporate aim on profit is predominantly skewed towards economic instead of social incentives. This alters the purpose of
measures taken by firms during extreme weather events, which influences consumers, employees, and inhabitants within disaster-struck regions. This portrays a situation in which mainly shareholders hold corporations accountable (Table 2). A coherent representation of a socio-ecological model could help in distinguishing the corporations driven predominantly by profit or social incentives, which may increase accessibility of useful opportunities during crisis management resulting from clear-defined goals and relations. The four concepts have existed longer within the academic literature. However, according to Comfort (2007), they must be understood in a new manner. He redefines the crisis management system as a complex and adaptive system by stressing the capacity to suddenly adapt to new conditions as crucial for effective performance. To achieve this capacity, cognition is of crucial importance. Baker and Refsgaard (2007) suggested multiple ideas similar to those of Comfort (2007) regarding crisis management, but are using the term disaster response instead. In correspondence with Comfort (2007), they emphasize adaptive capacity as crucial. However, they are adding concepts of complex systems theory like the earlier mentioned idea of resilience. By improving various components of crisis management, a system’s resilience increases.
Besides governance networks and their relations in the form of accountability structures, Hawkins & Maurer (2010) highlights the importance of social capital in describing the relations and dynamics of social networks between actors. With social capital, we mean the cultural capital in which the structure is made up of a set of social actors (such as individuals or organizations), sets of dyadic ties, and other social interactions between actors (Park & Miller, 2006). One such relationship is the bond between individuals and institutions or individuals who have relative power over them, such as employers and authorities. This then forms a broad concept which overlaps with the concept of accountability structures. It does however only encompass relations including an individual, and does not account for power relations between institutions. But, it is applicable to power relations outside governance networks. Hawkins & Maurer (2010) suggests that in a crisis situation residents, especially those with low incomes, rely heavily on social capital for individual, familial, and communal survival. Weak social ties increase vulnerability of inhabitants, as then the social ties with other families or institutions from a smaller ‘safety net’ to fall back on in comparison to when an individual has a lot of social ties. Fragile communal relations can result in enhanced dislocation when an ecological discontinuity occurs. Economic sectors collapse due to reduced occupation of jobs, demand and supply, which then forces regions further into downward spirals.
Another factor that is determinative for the consequences of a natural hazard and the effectiveness of adaptation measures is segregation. Knox & Marston (2016) define segregation as the spatial division of individuals based on ethnicity, race or class. According to Logan (2000), the standard measure of segregation is the Index of Dissimilarity (D). This is explained as the degree to which two demographic groups are evenly spread among neighbourhoods in a given city. Evenness is
defined with respect to the racial composition of the city as a whole. In other words, a 50/50 blacks/whites mix for a neighbourhood may still be regarded as separated if the citywide mix is 30/70. The index ranges from 0 to 100, portraying the percentage of one group who would have to move to achieve an even residential pattern - one where every neighbourhood replicates the group composition of the city. A value of 60% or above is considered very high, and values below 30% are considered low. Everything in between is considered moderate (Logan, 2000). Combining this index with the Gini Coefficient, a statistical measure used to assess income distribution among a population, results in a coherent synopsis of economic and social inequality (Shaughnessy, White, & Brendler, 2010). Segregation can be the result of governmental decisions and structures, and from the influence that businesses have in local communities as well. Socio-economic, ethnic and racial segregation are often the result of failures in long-term land use planning (Litman, 2006; Campanella, 2007). In the past, housing and zoning policies in cities often unintentionally - but sometimes intentionally - caused certain ethnic groups to settle in one area and other group elsewhere. This pattern often reinforced itself as time progressed, with members of one ethnic group settling closeby ‘hubs’ of their respective ethnicity, causing a higher level of D citywide, entailing a stronger degree of segregation. Racism and classicism in institutionalized forms, moreover, play important parts in weakening the relations based on linking and accountability structures (Stivers, 2007; Jurkiewicz, 2007). Lower of specific racial segments of the population can be seen as second class citizens, which results in failure of their citizen accountability. They are simply being less heard through the democratic process, which leads to low mutual trust between the people and the government.
Problem definition and complexity
The theories and concepts that are described, linked and integrated with each other demonstrate how adaptability consists of factors that transcend disciplinary boundaries. Figure 1 displays the overlap and relations between disciplinary concepts when integrated in a single framework as discussed above. The main problem that is tackled by this research is the lack of such an interdisciplinary framework that can be used for a comparison between New Orleans and Houston. Furthermore, the complexity that is involved with hazard adaptation becomes clearer by using such model. Each concept relates to a great variety of actors, which are highly interconnected and interdependent with each other. It forms a network of relations that is capable of and inclined to organising itself, and has the ability to learn and adapt. The theories and concepts describe crucial factors that are the result of and influencing the network in its aim to adapt itself to the occurrence of a natural hazard and to reduce the risk that is posed. The method of interdisciplinary integration contributed to better understand the structure and the functioning of this network and the relations between the determining factors.
Figure 1: Conceptual model resulting from the theoretical framework. Dotted lines indicate overlap between concepts; solid lines indicate a direct relation between the concepts.
Selected method and data
This research consists of a literature study based on existing qualitative and quantitative research regarding hurricanes and the adaptability of American cities. The qualitative literature research entices attention to the integration of theories by combining knowledge gathered from stand-alone studies and current developing theories (Atkins et al., 2008). Primary and secondary literature (e.g. academic literature, policy reports, etc.) are used …. Atkins et al. (2008) state that the synergy of such a qualitative approach with quantitative data could contribute to the identification of comprehensive and generalisable theory, add breadth and depth to existing theories by focusing on the persons that are directly influenced by adaptive measures, and determine insights into the arguments for failure and success of institutional interventions. Consequently, spatial analysis method is used to expose interdisciplinary interrelations.
Spatial Analysis Method
Research by Collins et al. (2010) states “that social data are increasingly becoming spatially explicit, which permits novel hypothesis testing and analysis that is spatially relevant, as well as multi-scaled” (p. 356). Using spatial data could connect monodisciplinary theories and lead to new levels of research collaboration, synthesis, and integration (Grimm et al., 2000). According to Joost et al. (2007), spatial analysis method (SAM) is founded on one of six concepts of spatial analysis researched by Goodchild (1996). To connect theoretical concepts from social disciplines (e.g. social segregation and accountability) with geophysical data (e.g. discharge flow rates, etc.), we turn the stand-alone input into useful information. Multiple visualisations, and the linkages between them, are used to reveal patterns and trends that otherwise might remain unnoticed under other methodological approaches. This approach associates information from multiple societal levels and makes interdisciplinary comparison possible via corresponding geographical coordinates (Joost et al., 2007). By means of applying the conceptual model on New Orleans and Houston, this study aims to locate societal bottlenecks during anticipatory and adaptive time spans. Color schemes are used to outline where these problematic elements are established within the two cities Combining the aforementioned methods could result in recommendations for institutional and social interventions, and programmes.
Results – New Orleans
The devastation caused by hurricane Katrina to the city of New Orleans in 2005 was more than just a coincidence. When evaluating the most important concepts and theories from the conceptual model (Figure 1), some results are obtrusive. Primarily, Colten et al. (2008) state that despite repeated
official warnings starting from the 1970s, the city of New Orleans remained highly vulnerable to hurricane strikes. Moreover, in the 5 years prior to Katrina, extensive warnings, with some of them even foretelling what would occur in Katrina, appeared in scientific papers, local newspapers and in disaster simulation exercises (Colten et al., 2008; Fischetti 2001; McQuaid and Schelefstein 2002; Laska 2004; FEMA 2004). Still the community and local government were not alarmed. This lacking attitude towards the high vulnerability of New Orleans concerning hurricanes caused an unprepared community and local government. This unpreparedness is also recognizable in the land use planning around New Orleans. Litman (2006) argues that proactive planning specifically failed in public transportation during the evacuation before Katrina hit New Orleans. According to Litman (2006), failures could have been prevented by better planning. Moreover, Litman (2006) indicates various long-term planning errors that contributed to the disaster of Katrina, and incremented the vulnerability of flood prone areas: the concentration of poor neighbourhoods in areas vulnerable to flooding, allowing shoreline development that eliminated protective barrier islands and wetlands, and underfunding levee maintenance (Bourne, 2004; Begley, 2005). The areas vulnerable to flooding are depicted on the elevation map (Figure 2) and in the actual flooding map of hurricane Katrina (Figure 3). An interesting fact is that the largest part of New Orleans is below sea level, this combined with the fact that the city is surrounded with water from large lakes and that it is located in the delta of the large Mississippi River, makes the city already vulnerable to flooding without hurricanes. Logically, the power of a category 3 hurricane has a devastating effect on infrastructure, housing and other properties. Land use planning could have helped protecting property of New Orleans citizens.
Further research by Burby (2006) describes how the government played a role in the anticipatory adaptation before hurricane Katrina. Burby (2006) addresses paradoxes in policy decisions across various levels of government that are related to poor land use planning and have contributed to the disaster in New Orleans. Part of the federal government’s policy was to create safe areas for economic and urban development in hazard prone areas by investing in adaptation measures and solutions for dealing with residual financial risk. However, the good intention to increase safety in hazardous areas in fact increased the probability of a catastrophe as it stimulated higher concentrations of the population in these relative high-risk areas, instead of stimulating concentrations in low-risk areas. This demonstrates clear failures in effective land use planning in the area of New Orleans.
Figure 2 (up) & 3 (below): elevation of New Orleans (m) and the actual flooding areas regarding Katrina, respectively (Swenson, 2013; Below et al., 2006).
Additionally, another element which failed during Hurricane Katrina was the governance network, responsible for preparation, response and recovery. The lack in performance of the network can mainly be traced back to numerous failures in the accountability regimes and the processes of crisis management (Koliba, Mills & Zia, 2011; Comfort, 2007). Koliba, Mills and Zia (2011) conclude that the governance network suffered from failures across all types of accountability structures. They have identified cases in which representative, citizen, market, legal, shareholder, consumer, bureaucratic, professional and collaborative accountability structures lacked in performance or broke down. This resulted in a system that failed to cope with the impact of Katrina on New Orleans. The
most apparent trade-offs occurred between bureaucratic and collaborative structures in the form of tensions between various levels of government and other non-governmental organisations. There was a lack of agreement on who had authority and responsibility for what functions. A striking example of failing bureaucratic accountability structures on the anticipatory side of adaptation is the paradoxes described by Burby (2006). The failure of the federal government’s safe development policy was reinforced by the passive attitude of the local government. It claimed not to be responsible for the safety of the area and the potential costs in case of a disaster, pushing it off to the federal government. This, while the federal government only limitedly covered potential costs, leading to citizens bearing the burden. There was a clear conflict on who could be held accountable in case of a disaster. Due to the incapability of major decision makers to deal with Katrina’s disruptions global brands conjointly invested more than $1 billion dollar in business aid within the first six months, while simultaneously cooperating with various disaster-relief organisations (Mishra, 2006). An increased emphasis on the market accountability structures and collaboration structures between non-governmental organisations arose as corporations stated the heightened demand for collaboration between the for-profit and nonprofit realms. According to Deryugina, Kawano, and Levitt (2014), such public statements and aid compensated for the loss of tangible and intangible resources and deterred harmful long-term economic effects. In the post-Katrina decade, CSR-related donations with respect to Katrina surpassed $4.25 billion, suggesting a societal emphasis on the recovery phase of New Orleans that was realized by shareholder accountability structures. Surprisingly, fast earnings recovery for victims hit directly resulted in higher post-Katrina wage incomes surpassing those of matched controls (Deryugina, Kawano & Levitt, 2014). Therefore, the rapidity of economic recovery is significant and stands in stark contrast with studies related to other large (non-ecological) discontinuities (e.g. Ruhm 1991; Jacobson et al. 1993; Neal 1995; Schoeni and Dardia 2003; Kodrzycki 2007; von Wachter et al. 2009; Couch and Placzek 2010; Walker 2013).
Lacking performance of the governance network was also for a large part related to bad functioning of the crisis management processes. Comfort (2007) states that the management of the situation in New Orleans was lacking cognition of the risk posed by the storm and its potential consequences. Scientists were aware of the potential danger, but government actors underestimated the information they received for reasons that are unclear. The result was that decision makers at all four levels of jurisdictional responsibility in emergency management failed to communicate an appropriate response to their respective agencies. Combined with a top-down governance approach, this led to a failure of the response system and to the response operations by private and non-profit organizations spiralling out of control. This resulted in the above mentioned shifting emphasis on collaborations between private and non-profit organisations. With direct and indirect costs estimates combined at ±$250 billion, supporting long-term recovery required federal and corporate actors to
increase accessibility to ordinarily elements for groups with low socioeconomic status. Lower income groups depend on access to investment funds, jobs, and safe housing locations (Masozera, Bailey, & Kerchner, 2007). However, conflicting incentives within accountability structures distorted societal recovery through misconceptions of what business actors deemed necessary and what actually was paramount for the disaster-struck groups. New Orleans passed a living wage ordinance, supported by 65% of the voters in the city and potentially raising wages of 50,000 workers, but collapsed under pressure from the business community (Masozera, Bailey, & Kerchner, 2007). Municipalities were restricted from setting fixed wage levels, which profaned the needs of inhabitants and clearly portrayed conflicts between the various accountability frames (e.g. market and democratic).
Another crucial aspect for explaining the disaster in New Orleans is the demographic layout of the population. In table 3 the demographic layout of the population of New Orleans in 2000 is displayed. The city remained the fourteenth most segregated city in the US, with D in the New Orleans metropolitan area as high as 75% (CensusScope, 2000). Both whites and Afro-Americans were impacted by the flooding caused by Katrina, but figures from Campanella (2007) illustrate that Afro-Americans were hit to a greater extent than whites, stating that Afro-Americans “outnumbered whites in the flooded area by over a 3.8-to-1 ratio (220,970 to 57,469). Blacks also outnumbered whites citywide before the storm, 2.4-to-1 (323,868 to 134,012)” (Campanella, 2007, p. 714). The strong segregated character of New Orleans is visualised, with the blue areas mainly populated by whites, and the yellow areas mainly settled by Afro-Americans (see Figure 4). Also displayed is the extent of flooding, with the red arrows illustrating levee breaches. Interestingly, most of these arrows point inwards to Afro-American communities, illustrating that those communities directly received floodwaters as the levees broke. Campanella (2007) states in his research that even though both ethnicities were impacted by the flooding, “there is no question that those who were stranded in the inundated city and suffered excruciatingly long delays in rescue were overwhelmingly African
American and poor (p. 714 -715).”
The segregated character of New Orleans illustrates the tense and polarized nature of the city of New Orleans in 2005, and is one of the reasons why little to no cooperation between ethnic groups and even within ethnic groups was identified (Forman, 2006). The higher impact of Katrina on the poor African-American community can in part be explained by the poor land use planning practices. New Orleans, a city in decline since the second half of the 20th century, was in a
problematic economic position. When the population of the city dropped, so did the city’s funding from the federal government. Little to no resources were allocated to in-depth hazard mapping, and levees which were supposed to protect the city were not maintained properly (Campanella, 2007). Moreover, according to Masozera, Bailey, and Kerchner (2007), wealth enables individuals to absorb and recover from social and economical losses more quickly. New Orleans’ median household income
($31,369) and per capita income ($19,711) were both situated under national averages, which, extrapolated to index D, suggests high poverty rates amid specific ethnic groups in society (Masozera, Bailey, & Kerchner, 2007).Post-Katrina Gini Coefficient for New Orleans is 0.5776, implying high income inequalities (Shaughnessy, 2010).Related to these economic disparities are daily-life factors resulting in disadvantages during and post-hurricane, such as lack of transportation, flood insurance, and sufficient financial credit regarding loan applications. Hence, the poor African-American
community was the most susceptible to these losses. Furthermore, impoverished Afro-Americans were possessing less social capital prior to Katrina in comparison to wealthy and white New Orleanians, and thus had most likely lower levels of risk awareness due to limited communication (Elliot, 2010; Hawkins, 2010). The lower levels of social capital means that less social ‘safety nets’ were available for the residents with less social capital, and consequently those residents had to rely more on themselves and direct family for their survival, thereby compromising other options for survival. Deficiency of such key elements steered the poor into a negative downward spiral from which it is difficult to withdraw. The segregated character of the city further intensified these factors, impeding cooperation with other demographic groups and government actors alike, leading to the pandemonium which followed in the aftermath of Hurricane Katrina (Forman, 2008). Moreover, studies have argued that institutional racism and classism undermined the citizen accountability of the poor and African American community (Stivers, 2007; Jurkiewicz, 2007). Because of the relative weakness, and social and environmental vulnerability of these segments of society, their ability to hold the government accountable was structurally undermined as they are perceived as second class citizens (Gomez & Wilson, 2008). This contributed to them benefitting the least from both
demonstrated that the more vulnerable were faster to blame the federal government for the failure of the governance system and its responses (Forgette, King & Dettrey, 2008). This may be an
indication of their lacking political authority towards the government and ephemeral dependance on financial aid from external actors.
New Orleans & Houston demographics
New Orleans (2000) Houston (2010)
White 28.1% 26.6%
Afro-American 67.3% 23.7%
Hispanics 3.1% 43.7%
Asians 1.5% 6.0%
Table 3: Demographics of New Orleans (2000) and Houston (2010) (US Census Bureau, 2012)
Results - Houston
The analysis concerning Houston’s responses and adaptive measures regarding hurricane Harvey is evaluated with the conceptual model in figure 1. Primarily, Houston ranks as the fourth most productive city in the entire U.S. in 2012, with a GDP of $489 billion in its greater metropolitan area (FRED, 2017). Its high-tech industry sector, paired with top-tier education institutions, result in plenty of capital and knowledge available in the city. Such positive socio economic factors should result in effective adaptation measures, but analysis shows irregularities. The first irregularity is the functioning of the warning systems and the awareness of the community and local government of Houston. Several news articles, and even recent reports, on hurricane Harvey suggest that Houston was aware of the impending hurricane due to well-functioning warning systems (Sebastian et al., 2017). Likewise, Gabbatt (2017) argues that Houston was well-prepared for hurricane Harvey, because flooding occurs often in the region. Sebastian et al. (2017) add that the local government had deliberately not arranged an evacuation plan considering many people already have flood resistant hiding systems in their homes. Although Houston seemed prepared for Harvey, the amount of rain during the six days Harvey has caused was unexpected. This rain caused more severe flooding due to inundation, consequently the more vulnerable areas of the city still had to be evacuated during and in the direct aftermath of the storm (Shultz & Galea, 2017). The flooding areas contain many rivers and flooding possibilities, which makes Houston vulnerable to flooding (Figure 5).
According to Gabbatt (2017), the vulnerability of the city increased in the last decade due to an expansion of new neighbourhoods. Drainage infrastructure has not kept up with the pace of the growing city, resulting in many neighborhoods still using roadside ditches as drainage (Gabbatt, 2017). Such observations shows Houston is not as well-prepared as was initially thought. In addition, decision-making in the Houston region is too short-minded, the long-term cumulative impact of decisions is often not considered (Gabbatt, 2017).
The main reason Gabbatt (2017) provides for the lack of long-term planning is the absence of coherence in the Houston government; metropolitan Houston is governed by nine different local county governments. Flaws in land use planning also appear in the Houston region. Impermeable roads are built on the coastal boundaries of the city, together with the flat topography of the region no run-off area for excess water is available. Furthermore, due to the extensive growth of the city, more high value properties of corporations and households have been placed at highly vulnerable locations (Sebastian et al., 2017). The effect of the hurricane on these capital-intensive locations is one of the main reasons for the damaging and costly nature of Harvey.
Closures of corporate facilities during Harvey’s extreme rainfall have again put emphasis on the vulnerability of such actors to ecological discontinuities. Research by Tan et al. (2017) states that Figure 5 (left) & 6 (right): Flood zones and total rainfall in the greater Houston area,
disruptive elements cascade through interdependent supply chain linkages that are distinctive of contemporary complex industrial systems. Total cost estimates of Harvey are $180 billion, which include damages to tangible and intangible resources throughout supply chains (Wile, 2017). Numerous organisations invested in disaster relief and financial aid related to Harvey; simultaneously building external and internal goodwill through actions exceeding superficial check writing (Yurieff, 2017). With companies sharing resources and funds with the American Red Cross, large shares of the destructions are tried to be compensated. Firms provide information about their CSR-initiatives through their marketing and communication branches, which make it relatively easy to pinpoint donations related to Harvey. However, given that contemporary instrumental CSR derives mainly from elaborate media coverage; which incentives are considered key remains difficult to decide upon. Economic rationale distorts ethical motives, increasing the difficulties of contemporary CSR analysis related to Harvey.
In response to data generated by forecasts prior to Harvey, crisis management authorities prepared the city by implementing multiple emergency protocols, such as the first decision on evacuation discussed above. Besides evacuation, resources and other protective measures were put in place and activated (Sebastian et al., 2017). As soon as President Trump acknowledged the situation as a disaster, the U.S. Department of Homeland Security and the Federal Emergency Management Agency (FEMA) gained control over federal emergency response. This was followed by the activation of several other national and regional institutions which supported in emergency preparedness measures and the spreading of information. A second decision was made to evacuate certain parts of Houston that became too dangerous, this time by FEMA. More than 15 other governmental organisations were installed and over 30,000 federal employees had been active in response, conducting over 122,000 rescues (Sebastian et al., 2017). More than 300 voluntary organizations, private and charity, were active for response to the situation. Throughout the response to Harvey, both before and after the impact, the coordinating emergency management authorities (most notably FEMA) worked closely with other agencies, including state and local organizations as well as non-governmental, community or volunteer organizations. The shared information, close coordination and collaboration not only ensured more resources were available, but that they also were used in a more coordinated and effective manner (Sebastian et al., 2017).
Based on the available (yet limited) information on the governance network responsible for the adaptive measures during Harvey in Houston, it may be concluded that this network performed relatively well. Both the established accountability structures and successful execution of the crisis management processes contributed to its effectiveness. Prior to and after the landfall of Harvey, the performance of the governance network seemed to reflect clear and well performing accountability structures. Good bureaucratic structures were reflected by the clear understanding of which
authorities were accountable for the response before and after the acknowledgement of the situation as a disaster by president Trump. This led to effective coordination, distribution of goods, communication and information diffusion. Moreover, the flexible conduct of the authorities led to the usage of a broad inflow of expertise from many different and also local sources. This can be viewed in terms of effective professional accountability. In terms of collaboration accountability, the system did fine as well. Once again, the flexible conduct of the governmental authorities made it possible for the many voluntary, private and community organisations and initiatives to be included in the decision-making process and to give them an amount of autonomy. Furthermore, shareholder accountability is reflected by the support of private companies in the area and the donations made as described above. Such actions depend on the agreement of shareholders. However, most private contributions to the management of the disaster had priority for clients and customers, which may be understood as an indirect usage of consumer accountability. This does demonstrate the limited inclusion of actors in the market accountability frame in the disadvantage of the more vulnerable segments of Houston’s society. Finally, as mentioned above, the incoherent government of Houston that caused flaws in land use planning may be interpreted as a failure in bureaucratic accountability structures. The different representatives of the Houston neighbourhoods called each other responsible for certain adaptation measures, which led to examples of lacking land use planning discussed above.
The processes of crisis management were also correctly executed. Before landfall, the dangers were timely identified, correctly interpreted, well communicated and adequately dealt with. Therefore, a sufficient level of preparedness was achieved. After landfall, the network was relatively flexible and adaptive to the situation, still able to stay coordinated and in control. This was because organisations did not have a specific emergency management focus, but were trained and prepared for such a situation so they could easily align their efforts. Moreover, new groups were formed from within communities that filled up or identified remaining gaps in the efforts of the governance network. Especially local and community groups have taken on themselves a crucial role in crisis management as they possess knowledge about the local environment and useful resources. The government seemed to adhere a more facilitating role instead of a top-down approach of crisis management.
Greater Houston is the most ethnically diverse metropolitan area in the U.S.A. (Kinder Institute, 2016). Even though the four main ethnicities are relatively evenly spread over the total population (see Table 3), ethnic and economic segregation is still prevalent in and around the Houston area (Figure 6). The Kinder Institute of Houston (2016) provide in figure 6 their results on ethnic segregation in the city. The darkest coloured spots on the map are the neighbourhoods in which over 90% of the total population constitutes of one ethnic group, either white, Afro-American or Hispanic (Kinder Institute, 2016). However, Index D dropped considerably over the past decade for
Houston, with that figure at the end of 2015 - the most recent data available - being a low 31% (FRED, 2017; Logan, 2000). It is striking to see how even though the D of Houston remains rather low, whites living in Houston are mainly situated in the southeast part of the city, and the other ethnicities primarily to the north, west and south. It is in those areas where most rainfall and subsequent flooding occurred (Figure 6). Nature is not selective, but data does suggest that in the areas other than the southeast, drainage systems and other flood mitigation measures have been maintained less strictly than elsewhere in Houston, mainly due to the older age of those neighbourhoods (Gabbatt, 2017). Julian et al. (2017) further quote the historical withholding of drainage system to communities of color as one of the primary reason those communities have worse functioning drainage systems, which subsequently played its part in the extensive flooding of those areas, which to this day often maintain their original demographic characteristics. According to the Kinder Institute (2016), these communities are situated in the western part of the city.
Median household income was $48,064 in 2015 and its normal distribution is skewed to the right under which the majority of ethnic minorities is located (StatisticalAtlas, 2015). Surprisingly, the Gini Coefficient of post-Harvey Houston is 0.5250, suggesting a relatively large range of income distribution between social groups (StatisticalAtlas, 2015). Wide economical gaps between the rich and poor imply a spatial mixture between high income and poverty-stricken social groups within Houston, as is shown in figure 6.
In a 2005 article, social systems architect and Houston resident Tory Gattis outlines a number of aspects from Houston’s local society that serve as benchmark for higher levels of social capital in the area. He quotes the more evenly distributed and relatively stable population demographic of Houston, affordable homeownership and the welcoming nature of Houstonians towards newcomers as the primary drivers responsible behind the rather high levels of social capital (Gattis, 2005). Higher levels of social capital generally lead to a more resilient social system (Hawkins, 2010). Elliots’ 2006 and 2010 research stresses the importance of social capital for individual, family, and communal survival; with higher levels of social capital, residents of the Houston area were able to cope with the effects of the hurricane more adequately than they would with lower levels of social capital. In an article by Pohl (2017), he further illustrates the high levels of social capital and thus strong resilience of Houston, as he states that during the disaster, Houstonians of all social, racial and ethnic groups, came together to help each other out. Not only were people who did not find shelter in their homes or refugee shelters taken in by others who were able to provide protection against the flooding, various restaurant also distributed meals and drinks.
Results - Comparison
In the analyses of New Orleans and Katrina similarities and differences occurred between both cases (see appendix A). The first similarity reflects on the historical vulnerability to flooding due to their geographical location in river deltas. Nonetheless, Houston has adapted better against flooding when compared to New Orleans, and thereby decreased its vulnerability. The preparedness in New Orleans was insufficient, even though they were warned on several occasions. Houston was better prepared and the society was warned for a heavy tropical storm in the form of hurricane Harvey. Similarities occur in the land use practices of both the New Orleans and Houston metropolitan areas. According to Litman (2006) and Gabbatt (2017) both cities lack in their long-term visions of proactive planning. In Houston the fast population growth was not covered adequately by authorities, resulting in poor decision making regarding risk perceptions, and therefore allowing society and corporate actors to locate themselves in areas prone to flooding. The drainage systems in large parts of Houston were not able to withstand large quantities of water. Consequently, the enormous and unexpected precipitation rates caused by hurricane Harvey resulted in inundation problems due to overloaded drainage systems. In New Orleans the more socio-economically vulnerable ethnic groups were located in the most vulnerable areas. These areas were often positioned near under maintained levees with an altitude below sea level, making these areas even more vulnerable to flooding. In both cities these errors could have been prevented by sufficient land use planning. Another difference between both cities is determined in the governance networks, which performed quite differently in Figure 6: Ethnic segregation in Houston (Kinder Institute, 2016).
both cases due to differences in the accountability structures and functioning of the crisis
management system. In case of New Orleans major flaws were detected in multiple accountability structures. Especially failures in bureaucratic and collaborative structures stood out as the main causes of unclear responsibility and authority of actors during the disaster. But, also prior to the storm, these unclear relations played a part in the failing land use planning processes. However, the governance network related to Houston performed better on these aspects, notably on the
bureaucratic and collaborative structures. Nonetheless, failures in land use planning were also found due to badly established bureaucratic structures.
The functioning of the crisis management processes in New Orleans functioned very poorly in comparison to those of Houston. Incoming data generated by forecasts were underestimated, which led to a failure in setting up appropriate communication within the governance network. Combined with the rigid top-down structure that was maintained by the government and its agencies, this led to lacking coordination and control. Non-governmental actors were badly incorporated in the
governmental response and they independently and uncoordinatedly undertook action in the area. Again, significant differences were found in the case of Houston. Forecasts were rightly interpreted and communication found place efficiently. Moreover, a more flexible and facilitating role maintained by the governmental agencies created better working coordination and cooperation with
non-governmental organisations. Related to this cooperation is the link between the accountability framework and CSR. Extensive media attention, combined with instrumental organisational
opportunities, resulted in a generous trend during both weather events. CSR-related measures from regional, and more external, actors regarding Katrina contributed for a share to rapid economic recovery (Deryugina, Kawano & Levitt, 2014). Similar activities are uncovered during Harvey, but, with more knowledge available regarding the concept of CSR and its ambiguities, actual measures
(especially their communications) could potentially contribute to a similar rapid recovery. The demographic setting between both cities was very different. Primarily, the population of New Orleans was divided unevenly between black and whites, and was also highly segregated. In Houston, the population is spread more evenly over a variety of ethnic groups, and also held a lower degree of segregation. Consequently, not one specific local society in Houston was victimized by hurricane Harvey, contrary to New Orleans where the poor Afro-American population was hit disproportionally hard. In Houston, many social, racial and ethnic groups collaborated to ensure each others survival, in New Orleans, hardly any cooperation between and even within ethnic groups was identified.
Additionally, this difference in collaboration can be explained by difference in social capital, which was significantly higher for Houston compared to New Orleans. For both cases, marginalised groups did however find themselves locked in hazard prone areas. Primary reasons for this in New Orleans was the neglectfulness from government officials, whereas in Houston the main reason was due to
historical circumstances and a lack of proactive land use planning. The highly segregated character of New Orleans, and the fact that vulnerable demographic groups were located in hazard prone areas, is one of the primary reasons the death toll of Katrina was so high. Coupled with low levels of social capital, survival chances for the vulnerable groups in New Orleans were slim. In Houston, the lower degree of segregation and higher levels of social capital made for a more resilient and adaptable city, further illustrated by the lower death toll, which is less than 5% that of New Orleans. From this then follows that in order to make a city more resilient against hurricanes, segregation of any kind must be low, and levels of social capital as high as possible.
Conclusion & Discussion
As is shown in the previous sections, various disciplinary factors influence and interrelate with each other in shaping the role of urban adaptability. Such transdisciplinary linkages have been used in answering the question of how past adaptation to hurricane Katrina in New Orleans compares to current adaptation regarding hurricane Harvey in Houston from a interdisciplinary perspective. Application of the conceptual model shows that lessons regarding extreme weather events from New Orleans have only been partially implemented in the present-day case of Houston (Appendix A). Nevertheless, limitations in scope and time result in this conclusion not being sufficiently covering for all actors within the related socio-economic systems. Only four disciplines have been used during this study, presumably neglecting relevant concepts from other academic fields of study that do not find themselves within our research scope. In addition, disproportionate availability of academic research related to Katrina in comparison with a sheer lack of academic resources for Harvey increases the difficulty for a balanced analysis. Notwithstanding, some recommendations related to future research ought to be made.
Firstly, the links between a multitude of similar cases should be examined further. This could result in more relevant and comparable data regarding contemporary cases, which results in an extensive database containing data applicable to internationally comparable events. An increase in applicability of relatable data could mitigate case-specific outliers and other deviations that distort results. Secondly, academic breadth should be augmented by adding more scientific disciplines to the integrative equation. Linkages currently disregarded ought to be studied in more depth. For example, the field of communication sciences could add to the current debate on the notion of ‘fake news’ altering the effectiveness of crisis management, and potentially worsening it (Sebastian et al., 2017). Lastly, quality of the results would improve substantially when primary and secondary data would have been collected by the writers first-hand. Ideally, in-depth interviews with inhabitants from the disaster-struck regions could potentially add to the social and scientific relevance.
So, multiple improvements on institutional, academic, and socio-economical can and should be made in the nearby future. This research has shown that the combination of societal faults with physical elements, even within progressive Western cities, sheds a different light on the statement made by the American president: “Remember, the USA is the most resilient nation on earth, because we plan ahead.” (Trump, 2014). Though, this research does not completely denies this, it is clear that a lot can still be gained in the area of adaptation. If, and how, America suffices this proposed role will be determined in the (regrettably) nearby future.
References
Amadeo, K. (2017). Hurricane Harvey Facts, Damage and Costs. The Balance. Consulted on 18/12/17, from https://www.thebalance.com/hurricane-harvey-facts-damage-costs-4150087
Atkins, S., Lewin, S., Smith, H., Engel, M., Fretheim, A., & Volmink, J. (2008). Conducting a meta-ethnography of qualitative literature: lessons learnt. BMC medical research methodology, 8(1), 21.
Baker, D., & Refsgaard, K. (2007). Institutional development and scale matching in disaster response management. Ecological Economics, 63(2), 331-343.
Begley, S. (2005). Man-made mistakes increase devastation of ‘natural’ disasters. Wall Street Journal, B1.
Bourne Jr, J. K. (2004). Gone with the Water. National Geographic, 206(4), 88-105.
Buis, J. J., Post, G., & Visser, V. R. (2015). Academische vaardigheden voor interdisciplinaire studies. Amsterdam University Press.
Burby, R. J. (2006). Hurricane Katrina and the paradoxes of government disaster policy: Bringing about wise governmental decisions for hazardous areas. The Annals of the American Academy of Political and Social Science, 604(1), 171-191.
Burby, R. J., & Dalton, L. C. (1994). Plans can matter! The role of land use plans and state planning mandates in limiting the development of hazardous areas. Public administration review, 229-238.
Campanella, R. (2007). An ethnic geography of New Orleans. The Journal of American History, 4(3), 704-715.
CensusScope (2000). Segregation: Dissimilarity Indices. Consulted on 3/12/17, from http://www.censusscope.org/us/rank_dissimilarity_white_black.html
Chakraborty, J., Tobin, G. A., & Montz, B. E. (2005). Population evacuation: assessing spatial variability in geophysical risk and social vulnerability to natural hazards. Natural Hazards Review, 6(1), 23-33.
Collins, S. L., Carpenter, S. R., Swinton, S. M., Orenstein, D. E., Childers, D. L., Gragson, T. L., ... & Knapp, A. K. (2011). An integrated conceptual framework for long‐term social–ecological research. Frontiers in Ecology and the Environment, 9(6), 351-357.
Colten, C. E., Kates, R. W., & Laska, S. B. (2008). Community resilience: Lessons from new orleans and hurricane katrina. Community and Regional Resilience Initiative, Oak Ridge.
Comfort, L. K. (2007). Crisis management in hindsight: Cognition, communication, coordination, and control. Public Administration Review, 67(s1), 189-197.
Deryugina, T., Kawano, L., & Levitt, S. (2014). The economic impact of hurricane katrina on its victims:
evidence from individual tax returns (No. w20713). National Bureau of Economic Research.
Elliott, J. R., & Pais, J. (2006). Race, class, and Hurricane Katrina: Social differences in human responses to disaster. Social Science Research, 35(2), 295-321.
Elliott, J. R., Haney, T. J., & Sams Abiodun, P. (2010). Limits to social capital: Comparing network ‐ assistance in two New Orleans neighborhoods devastated by Hurricane Katrina. The
Sociological Quarterly, 51(4), 624-648.
Fankhauser, S., Smith, J. B., & Tol, R. S. (1999). Weathering climate change: some simple rules to guide adaptation decisions. Ecological economics, 30(1), 67-78.
Forgette, R., King, M., & Dettrey, B. (2008). Race, Hurricane Katrina, and government satisfaction: Examining the role of race in assessing blame. Publius: The Journal of Federalism, 38(4), 671-691. Forman, T. A., & Lewis, A. E. (2006). Racial apathy and Hurricane Katrina: The social anatomy of
prejudice in the post-civil rights era. Du Bois Review: Social Science Research on Race, 3(1), 175-202.
Fox, T., Ward, H., & Howard, B. (2002). Public sector roles in strengthening corporate social responsibility: A baseline study. Washington, DC: World Bank.
FRED (2017). Total Gross Domestic Product for Houston-The Woodlands-Sugar Land, TX. Consulted on 11/12/17, from; https://fred.stlouisfed.org/series/NGMP26420
FRED (2017). White to Non-White Racial Dissimilarity Index for Houston County, TX. Consulted on 3/12/17, from https://fred.stlouisfed.org/series/RACEDISPARITY048225
Gabbatt, A. (2017). What makes Houston so vulnerable to serious floods? The Guardian. Retrieved Gallopín, G. C. (2006). Linkages between vulnerability, resilience, and adaptive capacity. Global
environmental change, 16(3), 293-303. from https://www.theguardian.com/us-news/2017/aug/28/houston-harvey-risk-floods-analysis
Gattis, T. (2005). Houston’s Social Capital. Consluted on 3/12/17, from
http://houstonstrategies.blogspot.nl/2005/12/houstons-social-capital.html
Godschalk, D. R., Kaiser, E. J., & Berke, P. R. (1998). Integrating hazard mitigation and local land use planning. Cooperating with nature: Confronting natural hazards with land-use planning for
sustainable communities, 85-118.
Gomez, B. T., & Wilson, J. M. (2008). Political sophistication and attributions of blame in the wake of Hurricane Katrina. Publius: The Journal of Federalism, 38(4), 633-650.
Grimm NB, Grove JM, Pickett STA, and Redman CL. 2000. Integrated approaches to long-term studies of urban ecological systems. BioScience 50: 571–84.
Hawkins, R. L., & Maurer, K. (2010). Bonding, bridging and linking: how social capital operated in New Orleans following Hurricane Katrina. British Journal of Social Work, 40(6), 1777-1793.
Hewitt, K. (1983). The idea of calamity in a technocratic age. Interpretation of Calamity: From the
Viewpoint of Human Ecology. Allen & Unwinn, Boston, 3-32.
Household income in Houston, Texas (City) (2015). Retrieved from
https://statisticalatlas.com/place/Texas/Houston/Household-Income#figure/household-income-percentiles
Johnson, B. R., Connolly, E., & Carter, T. S. (2011). Corporate social responsibility: The role of fortune 100 companies in domestic and international natural disasters. Corporate Social Responsibility
and Environmental Management, 18(6), 352-369.
Joost, S., Bonin, A., Bruford, M. W., Després, L., Conord, C., Erhardt, G., & Taberlet, P. (2007). A spatial analysis method (SAM) to detect candidate loci for selection: towards a landscape genomics approach to adaptation. Molecular ecology, 16(18), 3955-3969.
Julian, E. K., Lott, A., McCain, D., Palay, C. (2017). Why Houston remains segregated. The Houston Chronicle. Consulted on 21/12/17, from http://www.houstonchronicle.com/local/gray-matters/article/Why-Houston-remains-segregated-10935311.php
Jurkiewicz, C. L., & O'Keefe, S. (2008). A letter to the President-Elect regarding Katrina. Public
Integrity, 11(1), 79-90.
Kinder Institute (2016) The Shifting City: Houston’s History of Unequal Racial Change. Consulted on 3/12/17, from
http://www.kinder.rice.edu/uploadedFiles/Kinder_Institute_for_Urban_Research/Programs/Di sparity/Final_Kinder_Racial_Change.pdf
Klein, J. T. (2008). Evaluation of interdisciplinary and transdisciplinary research: a literature review.
American journal of preventive medicine, 35(2), S116-S123.
Klijn, E. H., & Koppenjan, J. (2012). Governance network theory: past, present and future. Policy &
