Risk and Vulnerability Assessment A Comprehensive Approach

Transcription

1 Risk and Vulnerability Assessment A Comprehensive Approach Niru Nirupama York University, Toronto, Canada nirupama@yorku.ca Abstract Disaster risk and vulnerability assessment depends on various factors such as, appropriate theoretical approach, quality and adequacy of information gathered, and community perception - for accurate and meaningful results. Accounting for people s perception and partnering with them in the process leads to deeper understanding of community vulnerability, which in turn provides better assessment of disaster risk: R (risk) = H (hazard) x V (vulnerability). Risk assessment is essential for prioritisation of various risks in time and space in order to plan for future disaster mitigation measures and resource allocation. This paper offers an integrated approach for risk and vulnerability assessment that includes theoretical disaster models, quantitative risk assessment method, and a component representing people s perception. The two models namely, the pressure and release (PAR) model and the access to resources (ATR) by Wisner et al (2004) offer static and dynamic approaches respectively. The framework recommended in the PAR model is valuable for basic understanding of the progression of vulnerability (in a given time) through identification of root causes such as, limited access to power and resources; dynamic pressures such as, lack of training and education, urbanization and rapid population change; and unsafe conditions such as, dangerous locations. The ATR model complements the PAR model by expanding upon the dynamics of changing decisions, options, livelihood opportunities, budgets, available resources, and choices made by the population that is impacted by a disaster, both temporarily and spatially. The hazard risk and vulnerability assessment (HRVA) tool is a quantitative method to evaluate community risk profile. The representation of people s perception is proposed by introducing an additional factor cp (community perception) in the fundamental risk equation: R = H x (V x cp). Integration of above described concepts/methods provides a unique and comprehensive approach to evaluate disaster risk for communities around the world. Keywords: Disaster risk, Community perception, Vulnerability, Mitigation, Disaster management

2 1. Introduction Recent trend in disaster management research demonstrates that concrete preventive and mitigative measures to reduce disaster risk to the population are important aspects of effective disaster management. In this century alone, we have witnessed tremendous loss of life, property and degradation of the environment in events such as, the Indian Ocean tsunami, Hurricane Katrina, massive floods in the United States, BP oil spill in the Gulf of Mexico, toxic sludge deluge in Hungary, hurricane Igor in Canada, and landslides in Columbia. These events are constant reminder that disasters do not differentiate between developing and developed nations affecting significant number of people each year. Livelihoods were at stake after the 2004 tsunami in many countries including Indonesia, Thailand, India, and Sri Lanka for several years (Nirupama 2009). In Canada, repeated events of hurricanes are causing east-coast fishermen great concern (CBC news, September 25, 2010) over continuing their livelihood. Many times, socially and economically vulnerable people living in hazard-prone areas cannot afford adequate insurance (Nirupama 2009; Nirupama (in press)). Wisner et al. (2004) suggest that social processes determine unequal access to opportunities, and unequal exposure to hazards. Positive impact of community-based and community-owned disaster reduction initiatives have proved to be successful in many parts of the world (Wisner 2004; ISDR 2004; Chambers and Conway 1992; Hewitt 1997). There is growing demand for integrating disaster risk analysis into programs for promoting community resilience building and sustainable social systems (GTZ 2004; Mileti 1999; Alexander 2002; Birkmann 2006), including decision makers input. It is a widely accepted sentiment that secure and resilient people will constitute sustainable and healthy societies (Etkin and Haque 2003). The Canadian Standards Association (CSA 2008) also acknowledges that perception of potential risk by stakeholders is a key issue. Vulnerability and lack of coping capacity result from social processes and institutional constraints (Tierney 1999). Risk perception suffers from consistent biases that can play an important role in creating vulnerability (Slovic 2000). In part these biases appear to be related to simple heuristics (in the absence of empirical data) used by people to assess risk. Thus perceptions of risks are certainly significant in decision making. For example, technical experts and other stake holders could disagree about the best course of action; or differences could arise due to people coming from different cultures. Furthermore, the level of risk acceptance and perception of risk vary among individuals, as well as between regions, communities and cultures (Schneider et al. 2006; Smith 2004; Olanubi 2009; Whyte and Burton 1982). In brief, new premise in disaster management emphasises upon understanding of root causes of progression of social, physical, economic, and environmental vulnerabilities of people; potential hazards risk; identification, evaluation and assessment of risk; and accounting for stake holders perception. This paper is an attempt to provide an insight into risk and vulnerability assessment processes through integration of community perception with various theoretical concepts, methods and tools.

3 2. Theoretical Concepts and Methods 2.1. Pressure and Release (PAR) Model The PAR model (Figure 1) approach provides a framework to assess the progression of vulnerability and evaluate disaster risk. The framework assists with identification of root causes, dynamic pressures (that translate root causes into unsafe conditions), and unsafe conditions. The Progression of Vulnerability ROOT CAUSES Limited access to Power and Resources; Political and Economic Ideologies DYNAMIC PRESSURES Lack of: education and skills; local investments and markets, press freedom, ethical standards Major forces: population growth, urbanisation, debt, deforestation, environmental degradation UNSAFE CONDITIONS Physical: dangerous locations, unsafe buildings Economical: livelihoods at risk, low income Social: special groups at risk, lack of local institutions Institutional: inadequate disaster preparedness, prevalence of health issues among population DISASTER Risk = Hazards Vulnerability HAZARDS Hydrometeorological, geological, biophysical Figure 1: Pressure and Release Model (adapted from Wisner et al. 2004).

4 2.2. Access to Resources (ATR) Model The ATR model (Figure 2) facilitates an understanding of how people make decisions based on their ability to access the resources in time and space, in the face of a disaster. It is a dynamic model that looks at people s access profile based on which they seek likelihood opportunities, both prior to and after a disaster. The model identifies dynamics of changing decisions, options, budget, access profiles, and choices made by the impacted population. SOCIAL RELATIONS STRUCTURE OF DOMINATION Demographic changes in time Households: defined by appropriate criteria (Sectors/Geographic/Income) Access Profile (resources and assets): land, capital, tools, network, education Decisions: savings/investment, selling of assets, review of access profile, acquire education, change location etc. Income Opportunities and qualifications required to access those opportunities Household Choices about which and how many income opportunities can be availed Livelihood: sum of chosen income sources and mechanism Household Budget: based on livelihoods; determine deficit or surplus at different times of the year Figure 2: Access to Resources Model (adapted from Wisner et al. 2004) 2.3. The Hazard Risk and Vulnerability Assessment Method Risk is considered to be quantifiable using probabilities and consequences (Helm, 1996; Wisner et al., 2004; Smith, 2004; Green, 2004). It is noteworthy that the standard risk formula has undergone significant changes in past several decades. For example, an earlier version of the risk formula was put forward by Blaike et al. (1994) as: risk = hazard + vulnerability. HRVA

5 method (HRVA 2004) was developed by the Provincial Emergency Program, Ministry of Public Safety and Solicitor General, British Columbia, Canada. HRVA accounts for severity (Table 1) of disasters and likelihood (Table 2) of hazards. The Risk Index (RI) is estimated using the standard risk formula as given in Eq. (1). Eqs. (1) and (2) are similar in that hazard is same as likelihood and vulnerability is synonymous to impact or severity. R (risk) = H (Hazard) x V (Vulnerability) (1) Eq. 1 can also be represented as: Risk Index (RI) = P (likelihood) I (impact) (2) Table 1: HRVA model numerical ranking scheme (adapted from HRVA, 2004) for the severity (or impact) of disasters. Category Details Assessment Description Rank 0 4 Very low 1 1 Fatality 2 Injury Critical 3 Infrastructure and Resources Property 4 Damage includes homeless, missing e.g. Hospitals, schools, utilities, transportation Public, commercial, private 4 10 Low High Very high Very low Low High Very high 4 Temporary interruption Very low 1 Closure of few days Low 2 Loss of 50% of capability High 3 Permanent loss Very high 4 Minimal damage Very low 1 Localized damage Low 2 Localized & severe High 3 Widespread & severe damage Very high 4 green/park, Minimal damage Very low 1 5 Environmental Impact asbestos exposure, toxic Localized damage Low 2 Localized & severe High 3 releases Widespread & severe damage Very high 4 6 Economic and Social Impact Industries, businesses and employers Temporary impact Very low 1 Temporary & widespread Low 2 Extended & widespread High 3 Permanent impact Very high 4

6 Table 2: Likelihood ranking (modified from HRVA, 2004) Return period in years Measure of likelihood Rank 1 5 Very likely Likely Slight chance Unlikely 2 >100 Very unlikely 1 In Eqs. (1) and (2), the value of likelihood is based on the return period (function of probability) of the particular hazard in a particular region. Ranking scheme for likelihood is given in Table 2. The HRVA method assists both, emergency managers and public alike in making risk-based choices to address community vulnerabilities in order to build disasterresilient communities. The information on return period hence likelihood for various hazards and their impact (severity) can be found with many local/ provincial/ federal agencies. For example, in Canada, the federal agency, Public Safety Canada ( compiles historical information on disasters which have directly affected Canadian people. The database describes where and when a disaster occurred and who was affected. In addition, federal organisations such as, Environment Canada and Natural Resources Canada; provincial authorities with emergency management portfolio; and local conservation authorities in each city also collect disaster data. International organisations such as, Relief-Web International ( US Geological Survey ( National Geophysical Data Center ( EM-DAT ( and Munich re ( also provide detailed information on disasters around the world. 3. Risk Assessment Accounting for Community Perception Risk perception is characterised as the intuitive judgement of individuals and groups of risks in the context of limited and uncertain information (Slovic, 2000). Community perception of disaster risk, disaster impact, resilience, risk reduction measures, and role of public institutions in disaster preparedness is vital in successful development and implementation of disaster mitigation strategies, measures and policy. According to an ongoing study of the author, which involves a group of immigrant women in Toronto, Canada, it was observed that their ability to cope during emergencies and disasters is extremely poor. In addition, these women (special group at risk) felt that the mitigation measures in place were not adequate in terms of their needs not being met. They expressed their concern over lack of community participation in the process of disaster risk mitigation planning and priority decisions. In other words, understanding various perspectives of immigrants with different cultural backgrounds is necessary before contemplating evaluation of their needs in the face of crises. It is noteworthy that nearly all the participants of the study had low to medium levels of education, were unemployed, faced language barriers (could not follow if there was a notice to evacuate), had low annual household income, and relied on public transit for medical and other needs.

7 Providing appropriate and adequate support to vulnerable families is fundamental (Nirupama 2008; Armenakis and Nirupama 2009) for a strong nation. Also important is to identify specific interventions (thoughtful of culturally sensitive issues) vulnerable immigrant families are expecting to receive during emergencies (Olanubi 2009). Marginalized groups are often negatively affected in the aftermath of a hazardous event because inequity makes them largely invisible to aid agencies and governments in times of crisis. As a result they experience the greatest impoverishment, disability, and fatality after a disaster (ISDR 2004; GTZ 2004; ADRC 2005). There is importance in the idea of building resiliency (Twigg 2007; Stewart 2007) for vulnerable population, as it is a means for them to better equip themselves for emergencies. Various groups of people, who are especially exposed to different hazards and risks, happen to be economically challenged in most cases (HRVA 2004; Mathew and Kelly 2008). In order to build resilient communities, it is primary to identify, evaluate and assess people s vulnerabilities and risks they are exposed to. Social (minority groups, women, disabled, illiterate etc.), physical (exposure to hazardous situation), economic (low income, unemployment), and environmental (land degradation, toxic deluge) vulnerabilities must be identified and assessed in consultation with the very people that are the main stake holders. Some experts have tried inclusion of risk perceptions of stakeholders or lay people within a given social system (Raaijmakers et al., 2008; Whyte and Burton, 1982). A modified risk formula as shown in Eq. (3) was suggested by Whyte and Burton (1982). R = p L x (3) Where x (> 1) is people s perception that depends on number of factors, details of which are not clearly defined in the literature. Ferrier and Haque (2003) have presented a methodology to implement the Whyte and Burton (1982) method but they have used community perceptions of the likelihood of hazard occurrence rather than impact of disasters. Thus, there is a need for a clearly defined methodology that incorporates people s perception of the impact of disasters, hence Eq. (4) is proposed as below: R = H x (V x cp) (4) Where, R = risk; H = hazard or likelihood (or probability); V = vulnerability/impact/severity; and cp = community perception of the impact of disasters. A 5-point scale as shown in Table 3 is proposed to quantitatively describe the community perception component, cp. A community includes all concerned members of a community, neighbourhood, city or region. It is assumed that the community under consideration will have sufficient access to relevant information regarding past occurrences of hazards in the region as well as their impacts on people, property and the environment. The Pressure and release (PAR) framework (Figure 1) will have to be constructed for a particular case in point, which will feed into the Access to resources (ATR) model (Figure 2). The detailed input from the PAR model will include unsafe conditions such as, physical exposure, local economy scenario, and non-existent public institutions. The

8 construction of the ATR model will provide a clearer picture of the past disasters, their consequences in time and space in terms of changing dynamics among people, economy, opportunities, livelihood options, and budget decisions. The ATR framework will become an additional and important resource to the community so they can develop their perception regarding past disaster impacts. Any disruption in critical infrastructure (essential services) such as hydro, water supply, natural gas, communication, and hospitals may carry an additional weight in determining the impact of past events. Table 3: Scale to quantitatively determine the Community Perception (cp) Description of Criteria Very severe occurrence(s) with severe impact on large number of people, critical infrastructure, and/or the environment hence consenting to positively invest in protection and mitigation measures as a priority Severe occurrence(s) with severe impact on some people, few essential services, and/or the environment hence consenting to cautiously invest in hazard protection and disaster mitigation measures Somewhat severe occurrence(s) with major impact on few people and/or essential services, and/or the environment hence consenting to invest in hazard protection and disaster mitigation measures in selected areas Occurrence(s) with significant impact on some people, and/or essential services, and/or the environment hence consenting to somewhat invest in hazard protection and disaster mitigation measures not a priority Occurrence(s) with minor impact on some people, and/or essential services, and/or the environment hence consenting to invest in hazard protection and disaster mitigation measures Rank Application of the Proposed Method The proposed risk assessment method as described in Eq. (4), shown in Table 5, and it s comparison with conventional method as given in Eq. (2) (Table 4) are presented here. Historical hazards data of the province of Ontario, Canada has been used for the application. The impact given in Table 4 for each hazard type has been determined using historical data (PSC; Etkin and Haque 2003) on fatality, injury, property loss, disruption in essential services, extent (localised or widespread) of impact, magnitude, duration, and cost of damage. In addition, the PAR (root causes, dynamic pressures, and unsafe conditions) and the ATR (social relations and structure of domination) frameworks helped with determination of more realistic and holistic estimation of the impact of each disaster event. Scenarios of livelihood implications on affected population, their decisions regarding moving places, acquiring qualifications etc., in order to access new livelihood opportunities provided a sense of people s perception during those events. Finally, based on factual information, ranks were assigned for likelihood using a 5-point scale as shown in Table 2, impact using a 4-point scale as given in Table 1, and community perception using a 5-point scale as given in Table 3.

9 Table 4: Risk assessment using conventional approach, as given in Eq 2, using the Province of Ontario, Canada data (hazard information compiled from Public Safety Canada disaster database). SN Hazard Likelihoo d (1) Impact (2) Risk Index (RI) (1) (2)= (3) RI (%) (3) 20 * 100 =(4) 1 Winter storm Wildfire Land subsidence Tornado Epidemic/ pandemic Extreme heat Landslide Expansive soil Hurricane Earthquake Hail storm/ wind storm Flash flood from 3 4 snowmelt * max value of RI, based on max ranks of Likelihood =5 (Table 2) and Impact = 4 (Table 1) Table 5: Risk assessment using proposed approach of accounting for Community Perception, as given in Eq. 4 for the Province of Ontario, Canada. SN Hazard Likelihoo d (1) Impac t (2) Community Perception (cp) (3) Risk Index (RI cp ) (1) (2 3) (4) RI cp (%) (4) 100 * 10 0(5) 1 Winter storm Wildfire Land subsidence Tornado Epidemic/ pandemic Extreme heat Landslide Expansive soil Hurricane Earthquake Hail storm/ wind 3 1 storm 1 3 3

10 12 Flash flood from snowmelt * max value of RI cp - based on max ranks of Likelihood = 5, Impact = 4, and cp = 5 4. Analysis of Results As evident from Tables 4 and 5, the proposed risk assessment approach produces different results than using the conventional risk assessment approach. For demonstration purposes, the values of Community Perception are used from Table 3. For fair comparison, each type of Risk Index has been represented in percent value. Figure 3 illustrates a comparison between Risk Index (RI) using convention approach and Risk Index using the community perception approach (RI cp ), confirming that people s perception may influence risk evaluation, hence decisions on disaster mitigation measures for a given region. For example, in Figure 3, evaluated risk for tornadoes varies from 60% for conventional approach to 12% for the community perception approach. In real world, for a given community or region, details on community s physical location, economic standing and institutional support status will allow for accurate assigning of community perception ranking as proposed in Table 3. Prioritisation of risks and vulnerabilities according to the assessed values of risk using community perception approach (RI cp ) is essential, not only for budget and resource allocation, but also future disaster mitigation planning.

11 Figure 3: Risk assessment using conventional and proposed approaches. The numbers on the horizontal axis represent hazard as given in Tables 4 and Conclusion Risk and vulnerability assessment is a key component to disaster mitigation planning and policy recommendations. Risk assessment enables prioritisation of potential risks in a region, making it the most essential factor required for the purpose of budget and resource allocation. Various theoretical concepts, frameworks, models and tools such as, Pressure and release (PAR), Access to resources (ATR), and Hazard risk and vulnerability assessment (HRVA) are available for the evaluation of disaster risk and vulnerability. The vulnerability (or impact or severity of consequences) is derived from detailed information gathered on people s vulnerability - social, physical, economic and environmental, potential damage to people, property (including critical infrastructure), and the environment, disruption in necessary services (lifelines). The conventional approach for risk assessment is discussed in detail along with its application. Realizing the importance of accounting for community perception in determining risk and developing disaster mitigation strategies and measures, a comprehensive approach that is inclusive of community perception, has been proposed. The proposed approach, as well as the conventional approach are being applied to past disasters data of Ontario, Canada. It is demonstrated, through the working example that Risk Indices (also referred to as risk factor or risk value in some literature) with and without community perception component can be substantially different. Thus, for a successful disaster mitigation program, community participation and perception is recommended to be considered as an important factor. In conclusion, this study can have significant potential to influence both, disaster management practice and public policy involving vulnerable communities. References Alexander, D. (2002). Principles of emergency Planning and Management. Oxford University Press. Armenakis, C. and Nirupama, N. (2009). Vulnerability Assessment using GIS: Toronto Propane Explosion. In Proceedings CRHNet Symposium, Edmonton Nov Blaikie, P., Cannon, T., Davis, I. and Wisner, B. (1994). At Risk: Natural Hazards, People s Vulnerability and Disasters. 1st Edition. Routledge. 284p. Birkmann, J (Ed). (2006). Measuring Vulnerability to Natural Hazards: Towards Disaster Resilient Societies. United Nations University Press. 524p. Chambers, R. and Conway, G. (1992). Sustainable Rural Livelihoods: Practical Concepts for the 21 st Century. IDS Discussions Paper No Brighton, UK: IDS (Institute for Development Studies), University of Sussux. CSA (2008). Emergency Management and Business Continuity Program, Z , Canadian Standards Association, Canada. Green, C. (2004). The Evaluation of Vulnerability to Flooding, Disaster Prevention and Management, 13(4):

12 Etkin, D. and Haque E.C. (2003): An assessment of natural hazards and disasters in Canada, Kluwer GTZ (2004): Guidelines Risk Analysis a Basis for Disaster Risk Management, Federal Ministry for Economic Cooperation and Development, Eschborn, Germany. Helm, P. (1996). Integrated Risk Management for Natural and Technological Disasters. Tephra 15(1):4 13. Ferrier, N. and Haque, E. (2003). Hazards Risk Assessment Methodology for Emergency Managers: A Standardized Framework for Application, Natural Hazards, 28: Hewitt, K. (1997). Regions of Risk: A Geographical Introduction to Disasters, Person Education. 388p. HRVA, (accessed on February 10, 2011). ISDR (2004). Living with Risk: A global review of disaster reduction initiatives. United Nations. 429p. ( Mathew, A. and Kelly, K. (2008). Disaster Preparedness in Urban Immigrant Communities: Lessons Learned from Recent Catastrophic Events and Their Relevance to Latino and Asian Communities in Southern California, The Tomás Rivera Policy Institute University of Southern California, School of Policy, Planning and Development, Los Angeles, California. Mileti, D.S. (1999). Disasters by Design, Washington, DC: Joseph Henry Press. Nirupama, N. (2009). Socio-economic Implications based on Interviews with Fishermen following the Indian Ocean Tsunami, Natural Hazards, 48 (1), 1-9. Nirupama, N., (2008). Disaster Risk Management: Theory, Practice and Policy. Workshop Report: Public Safety Canada, Government of Canada, Ottawa, 149p. Nirupama, N., Disaster risk management, Invited chapter in Encyclopaedia of Natural Hazards, Peter Bobrowsky (ed.), Springer (In Press). Olanubi, T. (2009). Crisis Management and Psychosocial Intervention: Perception of Immigrants, Major Research Paper, MA in Disaster & Emergency Management, York University. PSC Public Safety Canada: (accessed on Feb 12, 2011). Raaijmakers, R., Krywkow, Jo rg, and Van der Veen, A. (2008). Flood Risk Perceptions and Spatial Multi-criteria Analysis: An Exploratory Research for Hazard Mitigation, Natural Hazards, 46: Schneider, Th., Basler, E., and Partner, A.G. (2006). A Delicate Issue in Risk Assessment. In RISK21 - Coping with Risks due to Natural Hazards in the 21st Century. Ammann, Dannenmann & Vulliet (Eds). Slovic, P. (2000). The Perception of Risk, London: Earthscan Publications, 473 pp. Smith, K. (2004). Environmental Hazards: Assessing Risk and Reducing Disaster. Routledge. 306p. Stewart, R.M. (2007). Community Perspectives of Flood Risk and Social Vulnerability Reduction: The Case of the Red River Basin. Doctoral Thesis, The Natural Resources Institute, The University of Manitoba. Tierney, K.J. (1999). Toward a Critical Sociology of Risk, Sociological Forum, 14 (2).

13 Twigg, J. (2007). Characteristics of a Disaster-resilient Community. DFID Disaster Risk Reduction Interagency Coordination Group. 39p. Whyte, A.V. and Burton, I. (1982). Perception of Risk in Canada. In Burton, I., Fowle, C.D. and McCullough, R.S. (eds), Living with Risk, pp University of Toronto, Canada. Wisner, B., Blaikie, P., Cannon, T. and I. Davis (2004). At Risk: Natural Hazards, People s Vulnerability and Disasters. 2nd Edition. Routledge. 471p.